The Internet of Things
is the giant network of interconnected devices from these devices are capable
of making decisions without any human intervention. Edureka welcome you to the full course session
on the internet of things and what follows is a series of fundamental concepts
on the internet of things which will help you
get started with IoT. But before we begin let's look at our agenda for today.
so we can start out with introducing you
to the concept of iot. We are going to discuss. Why do you need I will take what is iot of you benefits
of iot for biofuel features and finally the very important five layer
architecture of authority. Then you're going to discuss
Raspberry Pi in this section. We are going to talk
about why you need Raspberry Pi what does Raspberry Pi
of you Hardware specifications and the installation
of Rosman fire also, This module. We have a few demos
including the Raspberry Pi a few cents hats and
Raspberry Pi camera module. Next. We are going to discuss
the top seven projects in iot in this section. We're going to talk
about the best projects that we found in Rog next. We will talk about iot devices are going to talk
about various devices which came into Inception after the concept
of iot was introduced then we want to talk
about iot applications in this Section we
are going to discuss practical applications of iot in various domains such as Healthcare
security traffic governance. So on and so forth and finally
we are going to discuss all things career in Ironton. They're going to talk
about the job market the opportunity the salary
in different geographies and experience so on
and so forth, but that we come to the end of our agenda also going to take
up this time to subscribe to us and don't forget to hit that
Bell icon to double missile. Update from The Edge
Arabia YouTube channel. So without Much Ado, let's get started
what is internet of things now to help you understand
what is internet of things? Let's look at an example
of a mobile phones Forest. Okay. Our mobile phones
has GPS tracking. It also has mobile gyroscope
you have adaptive brightness, which gets adjusted based on the light falling on it
you have voice recognition. And you also have face detection which identifies who is
the user now again iPhone x is coming with face detection. So again, these are
a lot of features that are coming pre-built on the mobile asset and most of these have a common
interaction between them because let's say
one application can use all of these features. I can also have these features in itself being interacting
with each other. Let's say based
on a GPS location. My brightness could be adjusted
or based on the direction. My phone is being held
the brightness in itself can also The adjusted as well. So there are a lot of features but when they interact
with each other these features come together to bring
in a better system that anything that they can
provide individually as such that's what internet of things is it
basically is a platform where we can connect
every day things which are embedded
with either Electronics software or sensors to the internet and this in turn enables us
to collect as well as exchange between these things. Now when I say things it
can be anything and Nothing, let's say I have
an internet platform where and I can connect these things if I take the example
of my house, I can connect my lock. I can connect my AC. I can connect my light and all this can be managed
on the same platform since I have a platform so I can also connect
my car to this. I can keep a track
of my fuel meter. I can keep a track
of my speed limit. I can also keep track
of the location of the car as well. Now if there is
a collective platform where all of these are connected
wouldn't it be great because I would love to Of the AC on and set
a cool temperature at my home by the time
I reach back from office if I have a platform that knows my preference and
that keeps track of where I am and where I'm going
to then it can also identify that I'm going from work back to Hope
and my preference suggest that it would be best if there was an easy
temperature of about 22 or 23 degree centigrade
and this is something that is definitely possible
through internet of things. Now, this is just one
of the examples now, let's say tomorrow
you come back. Home, it would be great. If I wouldn't even need
a key to unlock my door my home system should be aware that I have come home
and should unlock all the doors that are needed. And now this can be done if my mobile and my home
devices are connected onto the same platform based
on the location of my mobile. It can identify that I am at my hope
so it will automatically unlock the door and let
me come in as well. These are some
of the real world implications of Internet of Things. These are something
that are already happening but going forward what Need to understand is that when I have
a specific component with me, which can do a lot individually
wouldn't it be great if I can collaborate
this complaint with my system of different components
and build up at a system. This is what internet
of things is helping us to okay, you provide a platform to which
all these things are connected through the internet. So internet becomes the medium
through which you're connecting all these components were
things to a platform. Moving forward let's
try to understand. Why do we need internet of things now to help
you understand why let's look at an example. This is a patient at home. Okay, he's on constant
life-support wearing. His status is being checked to a health monitoring system
present on the cloud. Let's say at a point
there is certain issue with respect to his health. Let's say there's some
irregularity with his heartbeat or his blood pressure is low. There's some fluid being
developed or so far now what? Offenses since the system
on the cloud is connected to a hospital as well. This information would
get passed on to the hospital as well. They're in what would happen is that they would get
the complete details with respect to the patient and the important
information with respect to the current situation
of the patient as well. We made it aware with respect
to what issue exactly is the patient facing
as well as enable them to dispatch an ambulance immediately to bring the patient
back to the hospital as well. Now meanwhile, once the patient
has been picked up and brought back to the hospital
there could be prescriptions. There could be medicine. They could also be
an operating theater made ready in case of an emergency
situation as well. There'd be doctors on standby who have the complete
history of the patient who have the complete details of the present condition
of the patient as well. So this in turn brings
in a lot of transparency and reduces a lot of effort and time involved with respect
to this same thing. Let's take in today's scenario. There has to be someone
wandering this patient's health. If there's a fluctuation they
need to call the hospital they need to call request
for an ambulance. And meanwhile,
once the ambulance is here. They take the patient
and their back into the hospital then again there needs
to be a lot of checkups that need to be done because the doctors are
not fully aware again. There is a lot of tests
that need to be run and this in turn leads
a lot of delay as well in certain emergency cases as such if a system can do
this then this is exactly where our future lies in and what we have to definitely
move forward to to now Internet of Things basically
is expanding the interdependence of humans to interact contribute and collaborate with
things around us. Now. What do I mean by
interdependence of human see how we depend on each other like how I'm explaining
this concept to do if you're not clear you
would reach back to me. You would request me for help. Okay, this becomes
an interdependence between us when I'm giving you knowledge when you're not here
you're coming back to me and you're helping
similarly tomorrow if there's something
that you can do for me, I would reach out to you. We all are interdependent
on each other for something or the other if we can expand this interdependence
to interact collaborate and contribute with respect to
the different things around us. Then we would be building
a proper Internet of Things environment. This would be a much more
safer secure effortless and time-saving environment
into existence moving forward. Let's Talk about the various
benefits of Internet of things. Now. The first thing that would be as a benefit of having an Internet
of Things platform would that you could efficiently
utilize the resources that are available if I have a smart system
which can interact with everything if it has
enough computational power if it has enough understanding of how things work
between each other. I'm quite sure the usage of the resources available
will be more efficient as well. This resource could be
in terms of monetary. It could be in terms
of natural resource it. Also be an input taken up by the thing as
an input and so far. Okay. So all this can
be more efficient if I have a platform which is more smarter
and interconnected as well apart from this it minimizes
the human effort involved if my system is smart enough
to interact it to filter smart enough to do things that I do need to get involved with then my interaction
is always going to be minimum. This is same with everyone and
that's one of the major reasons why Internet of Things
has become popular today. The and the concept of smart home is always growing
as well in the same perspective because if the system in itself
is able to do most of my work at hope then I don't need
to put in much effort. I can relax at home without
having to worry about anything. Okay, the next benefit
would be it saves time. If it rains my human effort
definitely it is going to save my time. Okay apart from that if the resources are utilized
more efficiently then again, it is going to save
a lot more time as well. All-in-all any benefit
of Internet of things will in turn help you save
a lot of time as well. Okay. So time is one
of the major factor that can be saved on an Internet
of Things platform. Now if I have an artificial
intelligence platform through which all
of this is managed and maintained then
the personalization and the human touch also
comes into the picture. Now today most of us have had a level of interaction
with an artificial intelligence or a virtual
intelligence as well. This could be
a personal assistant like see Hurry, or it could be
an assistance application like Google assist now if I have system where all these components and things are interconnected then in turn all
the security present on each of these things
is going to get multiplied and it's going to build a much
more secure system apart from that the level of security that we would be integrating to the platform in itself
is going to be quite huge. So the overall security
booth respect to everything is going to increase
multiple times as well. Now let's look at some
of the major features of Internet of things. Now any technology that is available
today has not reached its hundred percent capabilities
and it always has that Gap to grow Internet
of Things is one of the major Technologies
in the world today that can help any other
technology reach its true and complete potential as well. Now there are mainly three
aspects to Internet of Things. Things as to how it works first
is the connect aspect here. Basically, what you need
to work on is you need to ensure that there is a connectivity
between all the things around you all the necessary
things to The Internet of Things platform. Okay, then comes analyze now,
I have my things around me. They each are going
to generate some amount of data now this data
needs to be collected and it needs to be
analyzed to build a business intelligence solution if I have a good insight, The data that is gathered
from all of this then definitely I can call
my system as a smart system. Finally what happens is
in order to improvise and improve your system. You need to integrate it with various models to improve
the users experience as well. Let's say there's
a personalization module that is there or let's say
there's going to be an aspect when I can directly
connect to my providers. Now. This provider could be Amazon. It could be Flipkart. It could also be my retail store
that is next to me. Let's say I'm out of milk to retail store
will get a notification and he would send
milk right away and the same thing can be done. Let's say I have my coffee
machine requires special beans and this be isn't is
almost complete then what my system could do
is it could go online to Amazon and purchase
these beans for me as well. So there are a lot
of things that I can do when I integrate this
with respect to various models and improve the overall
experience to a single user or a group of users
on the world itself such. Okay. Now, let's talk
about each one of One by one let's start with connect. Now. The first stage of connect
is device virtualization because what you
need to First do is that you need to standardize
the integration of the device to the Enterprise platform, which is present on the flower. Okay. Now, it could be present
on a cloud it could be present on a so, but again, it's all going to be connected
through the internet. So what I need to do is
that I need to ensure that certain level of standard
is present on the device so that it can go on and connect to my Internet
of Things platform. Now to help you
understand this better, there could be a standard
power plug and there could be a power block which has
an inbuilt Wi-Fi support so that it can
connect to my lap. No to build a smart home system. I need the second power point because there only I would have
the access to control it over the internet and my system could integrate
with respect to the same. Okay. Now if I take the first
standard PowerPoint then what would happen is that I need to manually
switch it on and switch it on how in the second case
I could send a signal to it and In don't we'll switch it on
and switch it off. So there's supposed to be
a level of standardization through which I can integrate
all of these devices to my platform. Next comes high-speed messaging. So now what I have done is that I have connected all
these devices to my platform but these devices in turn
generate a lot of data and this data is what is going to help
us understand better on how we can improve
the overall system and help and provide the user
with better experience. So for that we need
to To have high-speed messaging. Okay. This basically means that there needs to be
a reliable secure and a bidirectional
communication Channel between the devices and the platform not the purpose
of it being bidirectional is because you need to control
each one of them as well. Let's say I want to switch
on the AC then the signal would be going on
from the cloud platform to the device. So this is how it works on. Okay. So every communication needs to be reliable
it needs to be secure and it needs to be
bi-directional as well. Moving on to the third point of connect you need
to have endpoint management. If I don't have
an endpoint manager, I have established a way through which all my devices
can connect to my platform. I also ensured that the data
is going to be sent from the device to the cloud and the cloud can send back
to the device as well through a secure Channel, but if I don't actually identify from which device
which data is coming and how this data
has to be processed. Then it becomes a failure
of the system. This is where Management comes into picture endpoint management
basically helps you in managing the devices
endpoint identity the metadata and the overall life cycle
involved with respect to these things are such. Okay, so to put it quite simply
it basically helps you identify from which device
which data is coming and what needs to be done
with this data as well. Now coming on to the next
feature is analyzed and the first thing that you
need to do for analysis, is this reprocessing now if the data coming from the
device Not on a real-time basis, then my system
is often Williams. There's no use if I tell my system to switch
on the AC at my home. And by the time I reach there if the AC is not even
turned on then it's a failure on my system. Okay, so real-time analysis of the incoming and outgoing
data and must be done. I have with respect
to different aggregations it filtering correlations
processing and so far. Okay now apart from this
what you need to do is that this is raw data that is being streamed
from all the things you need. Need to identify which is contextually
important information which is going
to be taken forward. So once I have
the irrelevant information, then I can even generate
composite streams of information which can be taken ahead
for future analysis and understanding as well. This is what your data
enrichment process does. Then you have events
to now in events to basically any information that you want can be queried and visualized from
the vast amount of data which is present on my cloud. Okay. Now this in turn
can also help me get a better insight and Analysis if I have all the enriched data
present on my cloud platform. I have a tool which helps me identify
what is needed help me analyze. This data helps me visualize
it then definitely it becomes more useful as well. And when I have data coming from different things as such
this in turn can also lead to being a collection
of big data now when I talk about Big Data, it's not just a few GB of data as it's going to be
terabytes of data because because the data generated
from the things around us is that was and if you doing it
over a period of time then definitely it is going to grow
into a big data domain as well. Now coming to the third feature of Internet of Things
comes Enterprise connectivity. So this is what I
was basically telling you about. Let's say I have a requirement from my retailer or even
an Enterprise organization which is present. It could be Amazon. It could be flipcard anything as
as any Enterprise organization which provides me a service. If I can connect to them
through this platform, then definitely My
overall process also becomes easier as well. Let's say this is
service provider. Okay, let's say there is
a leakage in my Plumbing or let's say there's some issue
with respect to my electricity. Then it can contact to the
corresponding service provider. It can send them a detail and corresponding they
would be dispatched. This would in turn
reduce my effort of having to check the problem
having to call someone wait for them to come back all that gets reduced to minimum
required effort as well. Well, now how does
this communication happen? So for that we have rest API. Okay. Once I've integrated
my rest API with respect to the cloud application and my internet of things
then communication between the Enterprise communication
between the platform and the communication
between the things around us can be made more efficient and can be more easy as per the third aspect
is command and control if I don't have command and control on my platform
then it's awful noise. Yes, I built a very great. Table something that's
quite extraordinary. But if you cannot command it if I cannot control it as
per my requirement, then the system
in itself is not useful if I cannot tell my door
in a smart home to unlock when I want it then it
becomes failure on my part. If I cannot control the AC
on my horse motto then again, it's a failure. So always the major aspect when you integrate
with respect to these things. The major thing is that you need to ensure there's
a huge control on system and you're able to Um and it as per your requirement
now this command could either be through a voice
based recognition. It could also be a message that you can send
through your mobile application and so forth as well when we come down
to the iot ecosystem. There's no single
consensus or again. There's no singles
architectural design that's out there
which is agreed universally because each company
each organization each user for that matter has
different requirements and We look down to it. We can break it down to a simple
three level architecture where and we have
a perception layer where centers actually
gather the information from the environment around it. Okay. Once this is done
I'm going to use I'm going to pass this information
to the network layer the network layer in itself
takes up the responsibility of transferring this data from the sensors
to the next layer, which is the application layer
now here the main objective of the responsibility
is the application in itself delivers
this information to the end user or the When platform for
that matter this architecture can also be expanded
to a five layer architecture. Now when I talk
about a five layer architecture, it's quite similar here. It's a the difference mostly
comes around with respect to the transport layer
the processing layer and the business
layer mostly here when we had three layers
doing the earlier task. We've just broken this down so that we have
an easier operation or a smoother system
for that matter. Now again, the perception
layer remains same when it gathers the information
from the sensors, but the transport layer
actually Transports the data between the sensor
to the processing center. Now, this could be
through a wireless system. It could be through Bluetooth. It could be two RFID
3G NFC or any medium that I choose to once. The information has been
transmitted the processing layer comes into picture which actually stores the
relevant information analyzes this and again processes it as
per the users requirement. Now again, this could
employ various databases Cloud Computing Services as well as big data
processing modules to store this information as as well as process
it for that matter. Once this is done. I give the information
to the application layer, which is actually responsible for delivering Superior Services
to the end user for that matter on top of all of this stands your business lamp now
any device for that matter when it is working
on a large-scale environment a business layer
is usually used here. Now, let's say I'm working
in an organization where we using multiple pumps for different views features
in different locations for that matter a business layer
here actually monitors the complete functioning
of these parts. You can also have these
in various cars as well. So what do I would note here is if a car is going to break
down then I also get an awareness with respect
to that individual car and it also helps me enable or help me reach out to
the closest customer care center so that it can assist
the user coming down to how I can process it again. This can be divided
into two segments. I have my cloud computing
based processing. We're in here. It's quite simple
once I have the information. I pass it on
to the cloud platform which then in turn also. Processes it and also has
various applications to deal with this process information
for that matter. Now again, this is something
that I can do on a system which does not require
any immediate action and requires a large amount
of processing for that segment. But let's say I am in a system where I need immediate
response in those cases, I can go with my for computing now again
for computing is something that represents a layered
approach wherein we actually insert monitoring pre processing and storage with
the security layer. Between the physical
and the transport lab. Let me just go back
a few slides here to help you understand this. Now. If you actually look my for composedly comes
between these two layers when I add four new layers
for that matter. Now again, this is used in order
to make a system quite smarter or effective with respect to it
now between my physical layer and my transport lab, I have a monitoring layer. I have a pre-processing layer
a storage layer and a security line now to help
you understand this. Let's take in real-world
example out them. Let's say I have
a complete traffic system. Which is built
on my internet of things. Now, let's say at one point. There is an ambulance that has come to a first signal
I detect this and what I do is that I allow the ambulance
to move from this traffic signal by giving it a green line, but what you need
to understand is if I'm using a cloud computing this message has
to be passed on to the cloud. This information has
to be then process and then correspondingly a map
has all stood to be created at the same time when I use a fog approach
what happens is that all the pre-processing and the storage happens
on the Gateway level itself. The information from the sensor
goes all the way to the Gateway there and it actually processes this
stores the relevant information and sends this back to the discourage
ponding sensors as well. So let's say if there's
an immediate track that I can create
to the closest hospital. I would highlight all
the traffic signals to be green so that the ambulance can move
smoothly as well now security here is very essential because if I implement
this tomorrow anyone can actually try to manipulate this
for their requirements as well. Imagine there is a high High
Speed police chase happening and the culprit uses this in order to move fast
from the traffic signal. So this is something
that is really Ascension. Although there is a pro to this. There's also a con and this is how we try to overcome
this issue now talking about each one
of these new layers when I come
to the monitoring layer. What it actually does is it
monitors the power consumption it monitors each of these resources as
well as their response and the services that are running
on these resources. Now this in turn
helps me monitor or gives me a complete idea. Of which other services are which are the sensors
which are working? Where are the challenges? What is the power consumption
and how it works with respect to that same now once I have information
from these sensors what you need to understand is that usually you work with thousands of sensors
in a real-world environment. Now, I need to understand which
are the necessary information. So I'm going to do a level of filtering I'm going
to do a level of processing and then I'm going
to apply a level of analytics to understand what is needed and what is not from
these information as well now. Now the temporary storage area
is something that I use in case I want to store
any relevant information. Let's say I'm creating a route
today for an ambulance as well. And this is going to be stored
in my temporary storage area. But this also needs to be used
in future scenarios. So once I am done
with this usage, I can also push it
onto the next transport layer which can send it
to any other storage system that has part of my environment. Now as I said security plays
a very important role, although my for
computing is something that makes my system faster. It should not be easily. Inaudible now in
a cloud-based system. I have the assurance that it's not really easy
to break down the security. It's quite hard where there are
various layers of security which are part of the system. But when I have a fork system, it's essential that
this factor or the security which deals with the encryption which deals with the privacy
of the information. The Integrity of this
information is maintained. Now, there's also a very
interesting variation of this which is called an edge
Computing system wherein rather than doing all
these operations are After I have gathered
this on the Gateway, I can do it
on the individual notes or individual sensors as well
with respect to it. So where I have edges
these becomes point for me to perform
operations on the data that is being collected. So that's the slide variation of our for computing system
is such next let's talk about the various
taxonomy associated with internet of the X. Now. These are the key Concepts
or these are the key layers which are present with respect
to most architectures. That's out there. As I said, this is a general
guide ER e Person on each system that's out. There requires its own level
of customization requires its own level of approach
to solve that problem. But these always remain
the fundamental layers which are included in all the architectures
out there first. We have the perception layer, which is usually the layer where we gather the information
from the various sensors that's out there or we
use the various sensors which are required as
part of our system. Then we have
the processing layer where in we perform filtration. We summarize the data
we again do a level of analytics on This data before we decide to send
this relevant data to the system that's above this. Now. Then I have a communication line of communications layer
is very simple as in here will
Define the protocols and standards as
well as the medium through which the information
has to be passed from my sensors to my main system as well middleware is something
that's quite essential here. What it does is that it creates
an abstraction as well as it makes my system work
much more smoother. Now what you need
to understand is that there are various
components in what hear me? That really helps me integrate the information
coming out from each of these sensors or each
of these individual systems. Once it's present that I can pass it on
to my application layer where and I have various applications which help to improve
the overall experience of the user as well as provide much more accuracy and
efficiency to the information that's present now coming down. Let's talk about each layer
one by one now before I talk about
the perception layer what you need to understand is that one of the most
important aspects of Annette of things is context awareness. That is what you need
to understand with respect to the change of environment
is very important. And this is extremely impossible
without the usage of sensors as such now sensors in themselves are
very small in size. They again cost you very little and at the same time they
consume very little power again. There are various constraints with respect to the factors
as the battery capacity and the ease
of deployment as well, but let's not go
into them as such. Now when I talk
about sensors again, we have various types of sensors as such one of
the easiest example of sensors that can be seen
on a daily basis is the sensors which are part of your mobile. You have a location sensor you
have movement sensor camera in itself is actually another sensor your microphone
your light sensor. These are all various important
aspects of your mobile that we use on a daily basis apart from this neural
sensors medical sensors, like the fitness bands
that we use Healthcare bands, which are used
for heart patients. In environmental sensors, which check the temperatures
around the environment make you aware of the changes
chemical or biosensors which are very useful
on a daily basis as 12 again infrared
sensors are something that's quite common as well. When we talk about RFID. This is something
that's really important or not. This is something that really gathers a lot
of attention with respect to it RFID stands for radio frequency
identification now unlike a traditional barcode. It does not actually require
a line of sight of communication between the tags. And the reader
and can identify itself from a distance without even
a human intervention or a human operation for that matter RFID
is are technically of two types you have active
and passive active tags actually have some amount of power source
associated with it and passive sources do not have
anything related to it. And when we talk about
the RFID Technologies as well, there's near and far and near RFID reader uses
a coil through which we actually pass AC current
and Create a magnetic field. Now when we generate
a magnetic field anything that comes in its vicinity it
registers with respect to it. Now when I talk about afar
RFID it basically is a dipole antenna in a reader. Now this again propagates
an electromagnetic waves and tax themselves also have
a dipole antenna now again, these are something that's used
in various applications. That's out there. Now one key factor, which is associated with the perception layer
is an actuator now when I talk about an actual It
actually is a device which can affect a change
in the environment by converting any sort
of energy into another now. This could be a motor
which is generating electricity. This could be a windmill which is converting
the wind outside to you to of electric form as
well in these are just some of the examples that's out there and actuators themselves
play a very essential role in the perception lab now, the next layer that we have
is the pre-processing layer. But before I really talk
about the pre-processing layer, let's actually try to And the limitations of trying
it to process everything that we have
on the cloud system, which is part of our ecosystem. Now when I talk
about this one of the key or one of the biggest challenges
for me is mobility. Let's say my sensors are on devices which are
in constant motion. Then it becomes
a really high challenge for me to pass this information continuously
to my cloud environment again, this could be true the challenge
of Transport layer. This could also be due
to the challenge of power consumption
associated with it when my Smart device or when my sensors
for that matter are in constant motion
run constant Mobility. Then it cannot completely pass
all the relevant information on to the cloud. Now. This intern actually
causes a challenge for me to have some latency. This could also lead
to early 10 C with respect to real time processing
of the information that it gets as well. Now if I'm working
on a critical system, then real-time information is
something that I will highly depend on and that becomes
a challenge as well now. Now if I really want to scale up if I want to use a lot
of devices then my cloud computing system
also needs to scale but there's always a chance
to increase the latency because I'm working
with multiple sensors or multiple devices. That's out there. Imagine today. I have a system with just
includes thousands answers, but let's say in a smart
home system There is close to about 10 to 20 thousand
sensors associated with it. So this is just
one small system. Imagine if I'm trying to build
a smart city in that case. Is going to be hundreds of
thousands millions of sensors that's out there if my cloud computing system
cannot process this on a real-time. Then there's going
to be a high challenge with respect to that sick. And this is exactly where the usage of smart
gateways comes into a picture. This layer actually helps
me process my data on real type. It also helps me filter the data
on based on the priority or the requirement and creates a local copy
of whatever is needed or whatever needs
to be taken forward. Word now when I come
to the pre-processing layer or when I come to the features of for computing
for this matter, there's a very low latency because the information
does not have to go to the cloud system wherein the processing
needs to be taken up. It always is done
on the Gateway level itself. This information really
is faster as we've seen in the previous example
about the ambulance now, I can also use distributed nodes where in the information
does not have to be other processing has to be on one single node
for that matter when I'm using Distributed notes
then I can also distribute the effort or the work that is needed as well when I'm on a mobile
environment as well. These Smart Systems
can communicate with the gateways present
in its closest proximity. It does not have to connect
to just one single Gateway. If I'm setting up gateways
across multiple points in my city, then it makes it quite faster
and more efficient and this in turn can also lead
to a real-time response from the Gateway
for that matter. Once I have
a real-time response, then it is big. Taking my system faster and as in the previous example rather
than just clearing one signal. I can clear an entire path
for the ambulance as such now once I have relevant
information those which are necessary or those
which are really something that I need to ponder or I need to analyze can be sent
to the cloud system as and when as it's needed as such
so this is something that really makes my pre-processing layer
important and efficient and effective coming
to the next layer, which is the
most important layer. It's out that which is
my communication lab now as your Internet of Things
environment actually grows. This is a compromise or this is a combination
of various heterogeneous devices which are connected
to the internet. What you need to understand is that these devices
in then self need to pass these informations
and some of the challenges that the communication layer
should actually address is with respect to let's say
first start off with addressing and identifying of each of these informations
wherein I know which Each device is sending
me the information. What is that device if I want to communicate back
to that device as well. How do I do that again? When it comes down
to the communication in itself? This should also not cost me
a huge amount of consumption with respect to power because if I save up a lot of energy with respect to
how the information is gathered, but I waste a lot of energy
in transmitting this information then I don't have a smart system
again information itself, which comes should
use various route. Protocols which actually
require very low memory and should be very efficient
for that matter. If the information
itself needs to be bounced around different layers going
from different segments of your Echo System, then it needs to use
very less memory and this itself should be
very fast as well as seamless for that matter. Now when I come down to each of these components
for that segment, let's talk about NFC
or the near field communication. Now NFC is actually a very short-range
Wireless communication. Technology through which usually
mobiles interact with each other or a distance of a few
centimeters for that matter. Now all the type of data
can actually be transmitted between two NFC enabled devices in seconds by bringing
them close to each other. Now this in turn is actually
based on the RFID concept and it uses a variation of the magnetic field
to communicate data between two NFC enabled devices. If again, we go down into slide specifications and FC
usually works The frequency band of Thirteen point
five six megahertz. But again, this is very similar
to high frequency RF. Id I'm not going to bore
you more with respect to the technical details will
talk about the next segment, which is your RFID and wsn
integration for smart objects. Now again, many a times what you need to
understand is the data from one single sensor is actually not useful
for monitoring large areas and complex activities now here what you're going to use is that you're going to use
various sensor nodes to it. Interact with each other in this
also has to happen wirelessly. Now the disadvantage of a non IP technology such as
RFID NFC or Bluetooth is that it's range is very small, so they cannot be used in many applications wherein
large area needs to be monitored through many sensor nodes
deployed in various locations for that matter a wsn or a wireless sensor Network
consists of 10. Mm sensor nodes connected
using a wireless technology. They collect the data
about the environment and communicated to
the Gateway to Ice and relay the information to the cloud infrastructure
over the internet as such now when I come down to the iot network protocol
for that matter, usually what you
need to understand is that the IP for protocols
themselves can only be used for communication
of close to 20,000 device. Now again, the internet
protocols used by these devices is something that's quite a challenge as well
because when you look at it, the predicted amount of devices that's going to be available
by 2020 in The Internet of Things domain is
Close to 40 billion imagine 40 billion devices
communicating with each other. Now if I don't have
a smart system or if I don't have
an effective low power system, then I cannot communicate or I cannot gather
the information from this now usually are low-power IPv6
is used for these Network on these communication which helps you in passing of
information from these sensors onto your processing or onto your Cloud
infrastructure now again, when I come down to the low
energy technology my main, With respect to communication
is always to ensure that low energy technology
is most probably used in this segment. We have your
Bluetooth low energy, which is usually
referred to as a ble and this was actually developed by the Bluetooth
special interest group. Now what you need
to understand is that it actually has a shorter
range for communication and consumes lower energy as compared to
its Computing protocol. Now the BL a protocol stack
is actually quite similar to the stack used
in classic Bluetooth technology, however it As two parts, it has a controller
and it has a host as well. Now the physical and the link layer are
implemented in the controller and the controller
is typically an SOC or a system on chip
with a radio for communication. Now the functionality of the upper layers again
are included in the host and BLS actually not compatible
with classic Bluetooth. Now, the next is a low-power. Y5. Now. Again, the Wi-Fi Alliance has
recently developed a Wi-Fi Halo, which is based
on your IEEE 802.11. It had standards this
in turn consumes, very less power than compared
to your standard. Wi-Fi devices also
has a longer range. Now this exactly is why it is
most suitable for The Internet of Things applications
for that segment. Now any device that supports Wi-Fi also
supports IP connectivity, which is very important
for an iot application for that matter now. The last is ZB ZB is also based on the IEEE
802.15.4 Station protocol and is used mostly in
personal area networks or pan. Now again, the range for zigbee devices to communicate is very small
usually between 10 200 meters and the details of the network
and the application layers are also specified
by zigbee standards as such unlike the ble
the network layer which is part of the zigbee provides
for multi hopping routing the when I come down to more details
about this agree Network. I have three tabs which is an ffd full
functional device and RF. Reduce function device and
one zigbee coordinator as well. Now with this. I just hope you have
a simple understanding of how communication
is essential for The Internet of Things architecture and the various ways that you can implement
the communication between the devices as well. Now the next concept
or the next layer with respect to the iot architecture is
the middleware segment. Now when I come down
to the Middle where one of the key challenges and one
of the key issues that comes into the picture
is the interoperability. Operate ability as well as
the program abstraction imagine. I have 40,000 devices
communicating with each other 40,000 devices may not all use
the same programming language or may not pass the information
in the same way as well. I need to build or I need
to have something that ensures that these devices
communicate with each other and there is
an abstraction maintained between the information pass
from these as well. Now if I have
multiple devices also what I need to ensure is that these devices
Independently discoverable and I can manage
each of them today. I need to be aware. If one single sensor
also breaks down because the information
coming from the sensor is extremely important as well when it comes
down to scalability. It is extremely useful because when I need to grow
my ecosystem this middleware really comes into picture if I can replace
an existing middleware with something I
can help me scale up, then I don't have to completely
V8 my entire ecosystem as well with respect to it
usually When I use a highly capable middleware, then it also lets me perform
big data analytics and Implement security
and privacy as well and this in turn usually
helps me communicate with my cloud computing and also context
detection the again when you come down
to the middleware segment, you need to understand with respect to
the various specifications of the application. Which kind of database
is it oriented. What is the semantics
it's based on what kind of events can process and what kind of service
can it process or White ass but these are some aspects that you need to keep in mind while you're selecting
your middleware for your architecture. Now the last layer is
your application lab Now application is something
that really is what your end user gets or is what usually maintains or helps process
your information to the best that's out there. Now this in turn can be used
in different domains. It can help you achieve
different things. It can help you have
a smarter lifestyle it can help you have a smarter environment
your entire home system can be Managed with respect to an application
your car management can be done using application. You can build
an entire social life and entertainment system based
on a smart application now, although the end user
usually only looks at the application layer. This actually is the front face
of your entire architecture. Now with this. I hope you at least
have gotten an idea of what the entire skeleton of your iot architecture
comprises of now what we have discussed
here is just the skeleton. It's always up
to the user to add. Muscles to it and complete
it with a skin as per your requirement Raspberry
Pi today has become so common that you can find it across
most households as well today. It has become one
of the most cheapest and common Computing device that can be found
almost everywhere, but let's actually go back
to understand the ideology which bought Raspberry Pi
into development itself. Now Raspberry Pi basically was bought by the Raspberry
Pi Foundation to Introduced or to bring in the information
technology back to the schools where in students can learn how to program
from scratch the growth of technology today
has grown to such a level that everyone today has ease of access to do
anything on a computer but back in the early and late 90s and 2000 for using
a computer you needed to know how to program and how to work around
with respect to it. So this in turn have to build a very strong foundation
for programming knowledge And with the growth of UI
everything has become so easy that today you don't need
to learn programming to do much. But at the same time
this has made it harder for people to identify and understand good programmers
from those who are not. So in order to
build a generation which starts with a very
strong programming foundation and fundamentals. The main ideology
was to introduce or take back it to the basics and make it accessible
across every school as well. Again, as I said, it was introduced by
the Raspberry Pi foundation in 2012 as watching. The moving forward what
exactly is Raspberry Pi now. This is one of the questions
most of you would have a cross your mind as well because you
might have heard about Raspberry Pi body
might still not have a Clarity. So let me help you clarify
that right away. Now as per the definition
Raspberry Pi basically is a series of very
small single board computers, which actually have
additional features as Bluetooth Wi-Fi USB capabilities
General input Output ports and so forth now it basically is a very small
low-cost credit card computer which has actually
can be plugged into any monitor as well as
you can include a keyboard and mouse and it
increases the opportunity for people to explore
learn and understand how to program as well. Now the latest version of Raspberry Pi
is Raspberry Pi 3, which was released
in February 2016 as well. Now Raspberry Pi
basically is a combination of raspberry operating And Pike which basically stands
for Python programming language. But before we move forward, let me just show you
a very simple video which Raspberry Pi has put out so that you can understand the ideology of Raspberry
Pi Foundation as well. This is a Raspberry Pi it's
a credit card sized computer that costs around 25 pounds designed to teach
young people's program and is capable
of doing all kinds of wonderful things
back in the 80s kids had to learn
how to code computer too. You use them and as
a result these kids grew up with an inbuilt understanding
of how computers work. Now, we need more programmers than ever before so to deal
with this problem. Some clever people came up with the Raspberry Pi
to reignite the spark. It runs minutes
a free operating system from an SD card just like
the one in your digital camera and it's powered by
USB phone charger. You just plug in a mouse
and the keyboard. To a TV or Monitor and you're ready
to go in schools. Not only is best berry pie a great way to learn programming
skills as part of ICT. There are also dozens of cross-curricular applications
next science and music and all over the world. People are experimenting with Raspberry Pi's and
attending raspberry jam events where people of
all ages are learning what can be done
with a Raspberry Pi since the first
Raspberry Pi was ship. We've seen examples of people
using the pipe in a variety of amazing interesting projects taking advantage of its size
portability cost programmability and connect ability. And whether you want to learn
to make games build robot or even teach a bath parachute with Raspberry Pi
the sky's the limit. So I hope you guys had
a great learning experience with respect to that. That was basically the ideology
of the Raspberry Pi foundation on introducing Raspberry Pi
into the market as well moving on some
of the capabilities that Raspberry Pi
enables you to do is that it helps you to browse
the internet as well as what complete HD videos
on the same device as well. So all you need to have
is a HD supported display as such now apart from that even basic operations
like making spreadsheets. Creating words presentation
all these can be done on Raspberry Pi and you
have a huge set of games that are available which can be played
on Raspberry Pi making it quite interesting and easy as well for people
to enjoy the component as well. Then you have various
add-on capabilities, like infrared cameras
and security system, which can be built
keeping Raspberry Pi as the core Hardware as well. Then you can also use
many music machines as well as detection
of weather stations, which we actually had done
in our previous session. On as well. Now. These are just some
of the top capabilities that I have picked up
from the list Raspberry Pi today has become one
of the biggest component that has enabled
users to achieve and create a lot as well today. The capability of Raspberry Pi
is restricted just to your imagination whatever you
can imagine can be done using Raspberry Pi given the amount
of effort put into that. Mmm now moving on let's look
The Raspberry Pi Hardware as well before that. Let me show you how I would be using
a Raspberry Pi as part of this tutorial session. So let me just give
you a simple overview of the component as such now
what you basically see is that the Raspberry Pi Hardware. So let me just switch over the camera and let me show
you the Raspberry Pi now here what you're seeing is the
Raspberry Pi 3 Hardware as such and this is your processor
and system on chip, which is an adrenal
a few P3 processor. Now two important ports
here is the CSI. Which is the
camera serial interface where you can connect camera
directly to the Raspberry Pi and the DSi Port which is
the display serial input Port. Okay. Now, let me just flip
this over slightly. And if you see this
is a 2.5 M. M-- micro USB connector. So your standard mobile charger
can be used to charge the Raspberry Pi and this is a standard
HDMI port for display. So your standard
TVs and monitors which have an HDMI support
can be used to become a video interface for that. Then you have an audio jack, which is used to connect
your audio input now, let me just flip the slightly
over and let me zoom out a bit. Now if you see we have
an Ethernet connector following which there are
four USB slots as well. So when you see here, there's a 40 pin General
input output pin present here. So this is something that's quite interesting and configurable from
your Raspberry Pi a software where you can configure for what
each purpose pins can be used which will be silly little head. Now the final thing that I
want to show you is behind here when you flip backside, so there is a micro
USB slot present here. This is mainly to insert
your memory card. And when I say memory card, I would generally recommend
at least a 16 GB memory card. This is mainly because the Operating system
and itself is about 4 GB and if you use an AGP, then what happens is there is
very Less storage space for your operating system
to work on as well. So 16 GB memory card
would be really helpful. And one thing that you want to see is
this is an Evo class, which is something similar
to a class 10 memory card. So if you get a memory card, make sure it's at least a Class
A Door class 10 memory card so that you can have
high disk read and write processing
now coming back. Now bow your head let's just
skim through the various changes that has happened
over the hardware of Raspberry Pi now. We are not going to devote
too much time into this because this is a completely
in detail session how I would be stopping
at these slides for 10 seconds. So if you wish to know more
you can pause the video and get a complete look as well. talking about the processor
the first generation of Raspberry Pi initially came with a broad
Camby CM 2 8 3 5 SOC and it basically was similar to the first
generation smartphone chips and the architecture that was used also was
an armv6 architecture now the over the years
Raspberry Pi 3 has grown a lot and today it has a capability
of 1.2 gigahertz which basically is
because of the a arm cortex a53 64-bit processor it is
considered to be 10 times faster in comparison to Raspberry Pi wow now I'm talking
about the change in raspberry pi over the years as you can see here the various
models have been listed and the architecture change has also been mentioned
here as well so in case if you are interested in Rome
or you can pause the video and take a look at this now when you come to the memory
the first model of Raspberry Pi came with a 256 MB RAM and which basically
were shared by the GPU but today with Raspberry
Pi Version 2 and 3 you have four times that you have about 1GB of ram which again is shared
by the GPU as well. Now the default split was at hundred and ninety-two MB RAM
for the CPU basically, which was more than
enough to play a full HD that is 1080p video
or perform simple 3D operations, but again, not too
complicated operations as well. Now moving forward when you look at the networking
capabilities in terms of networking capabilities, the Model A of Raspberry Pi
did not have any such features, but from model B
of version 1 itself, you had an ethernet port which was introduced here
and from version 3. You also have Wi-Fi and Bluetooth capabilities
introduced here as well. Now talking about
the peripherals in Raspberry Pi there are basically
four USB ports. Now, this has been introduced in
model B version 1 plus onwards and today you have a lot more that you can do with respect to
that even audio input ports and audio output ports
have not Over the years but the on-board storage has changed a lot
with respect director. Now again, when you look at the video
capabilities video controllers, basically here you
can watch complete HD videos but Raspberry Pi does not have
a hatch 256 decoding Harbor but the CPU in itself is
way more capable and helps you in decoding
the h26 included videos through software operations now in terms of GPU the GPU in Raspberry Pi 3 runs
at a higher clock frequency, which is about 300 or 400 in comparison
to the previous versions which is at about
250 six megahertz. Now in terms of video input
you have a 15 put CSI connector that has always been present
with the Raspberry Pi and we output has changed
over the years. You have a HDMI port
you have a DSi port and now you also have
a 3.5 mmm trrs Jack as well. Now in terms
of the connector capabilities. There are 17 pins of the gpio port switch
can be configured as per your requirements as well. No, moving forward. These are the various
details with respect to the general input output pin. This data has been
taken from Wikipedia. So in case if you're looking
for more information, you can definitely
check out wiki space or the official Raspberry
Pi documentation as well now coming down. Let's begin the Raspberry Pi
installation process here. But first, let me help you understand
the different operating systems that are available at present. Okay Raspberry Pi
in itself supports. Has multiple operating systems
as such but we will mainly be working on raspbian. Okay. So these are some of the most
popular operating systems supported on Raspberry Pi so you have risc OS you have
FreeBSD operating system. You have net BSD operating
system Plan 9 is again from Bell Labs as well as
Windows has its own Windows 10 of iot version. Okay. So for this session, we're not going to go
into the Windows 10 will stick to the core Raspberry
Pi operating system now for downloading
the operating system. You can go on
to the original site. That is Raspberry Pi Dot o-- r-- g-- And there in the downloads tab, you can download
the operating system. Let me just show it to you. Okay. So this is the Raspberry Pi
home page and here if you go on to the download section, you can download
the operating system. Now if you are starting off
with Raspberry Pi an Internet of Things trust me. This is the best place
that you can find a lot of informations
with respect to it. Okay, most people actually published their details
with respect to the projects that they're working
many popular projects are also available. You can find a lot of help as
well as good information here. Okay. So this is definitely
the place to go to if you go on to the download section
here, okay this I'll show you 2 options you
have nodes or you have raspbian. I would recommend
that you go with notes because this is a complete
package with respect to the different operating
systems as such so here again, you have noobs and Loops like so
my recommendation would be you download notes and make sure you using a 32 GB memory card with this
would be really helpful because it gives a lot of memory
for your operating system. So in case if you're going for a 16gb or below I
would recommend you go with the nope slide, but make sure you keep the bare
minimum of 16 GB in case if you working with Raspberry Pi
but do not mistake me. When I say use
a 16 GB memory card. You can install it
on an 8GB as well. But again, the operating system
has less memory to work. So at least go for 16 or higher 32 is
what I would recommend. So once you download
this ZIP file, or you can even download
it by a torrent. Okay, but I would recommend
download it as a zip file. Let me show you the file. So this is the noobs file. Let me extract this. Okay, it may take some time. It's about a 1.5 GB file. Okay, the lights were seen as
slightly lower in comparison, but this is something
that I would recommend in case if you're working
with Raspberry Pi. Okay, so I have
my noobs folder here. So what you need to do is that you need to copy everything
inside the nude photos and then paste it
inside the SD card. Okay, make sure you are
not directly copy pasting the folder
that you have extract. So what happens is
if you do that, it does not actually recognize
the operating system as part of the Raspberry Pi
so copy everything that you've extracted
and then directly paste it inside the SD card. Now again, this
is why I recommend that you use a high
speed memory card because in case if you are not using it then what happens is
But it takes a lot of time with respect
to fetching the data. Okay. Now as I had mentioned if you're going for a memory card
for your Raspberry Pi you can use something
like a 16gb at least and make sure it has
a high transfer speed. So this in turn will give you
a lot of memory area to work around with as well. As I would recommend
you take something like a class 10 memory card that is available
in the market today. So these have a very good read
and write speed ratio. So definitely this is something
that you should consider while you're getting a memory
card for your operating system. Okay now Once you're done
copying with this, let's go on to install
this memory card onto a Raspberry Pi and let's
begin the installation process. Okay, so now we've
successfully copied all that is needed
for our operating system. So safely remove this make
sure you rejecting it safely because sometimes
what happens is if you do not see if rejected the
files get corrupted before you load it
into your Raspberry Pi this is certain issue with respect
to the memory cards. So just eject this. Okay. So this is my Raspberry Pi and
if you actually flip it over. This is the slot where you
have to insert the memory card. Now, we've copied
the noobs folder into this memory card. Okay, let me just insert this
into this and let's begin with respect to
the installation process. Let's begin installing
this memory card onto my Raspberry Pi
and start with the setup. So, let me just turn the
on my Raspberry Pi. Okay, so this is a screen. I'm getting by default. Now what happens is the
installation steps initializes. So the Raspberry Pi
is getting loaded. Now there are
two operating systems that are present here by default
and we want raspbian. So let me just select
raspbian and click on install option present here. If you see here it
basically is telling me that it requires 4.5 GPA. So this is why I was recommending you
to get at least 16 GB or more because if I have an 8 GB memory card if 4.5 GB goes off with
the operating system itself, then there's not
much memory left for me to work around with now. Let me just click on install. Just before I do that in case if you are using
a different language, or if you want to use
a different keyboard as well, you make sure you're
specifying that okay. So there are written language
is present here. So choose the
corresponding language as per your requirement. Once you selected
that description install and this is just
basically going to ask you whether you want
to format your data and install the operating system
on top of this. So click on yes, and the installation process has
begun now it may take some time. So I request you
to be a bit patient because this is an Operating
system installation process, so please be a bit patient
and hope you have fun mean wire. So now as you can see the
operating system installation has completed successfully
once you hit on okay, it's going to ask me whether I want to reboot or it
automatically reboot as well. Okay. So yeah, it's automatically
rebooting at this point. Okay. Now as you can see the Raspberry
Pi operating system has successfully been installed. Okay. So this is what your Raspberry
Pi operating system would be looking like okay. So this is raspbian
as such so there are different operating systems
that are available today for you to install that we
have discussed already now that we've installed
raspbian on our system. Let's look at some
of the accessories which can be used
with Raspberry Pi now one of the most popular and interesting accessory to
Raspberry Pi is the sense hat. Okay, so that basically is
a hardware attached on top. There are various hearts that are available
but censored is the most popular because you have various
onboard sensors present here. You have a temperature humidity and pressure sensor present here
as well as a gyroscope and a joystick also present
here apart from this. The reason that it
is quite popular is because it has an 8 cross
8 LED Matrix display that is present to him. So let me just switch over and show you the sense
hat component as well. And okay. So here you can see
the sensor know how it is slightly different from
the the other accessories is because it's got an 8 cross
Aid LED Matrix present here, which can be used
for displaying various things and we will also be seeing
how you can display this as well now coming on
to the other components here. Now if you see here, it's already got
multiple sensors here. It's got an accelerometer and the gyroscope
meter present here. It's also got we just
this part the axe low and gyrometer present here. So you can see it's got
labeling as well. Then you have
a humidity pressure and temperature sensor
on board with respect to this and it's quite interesting
and useful as well. Now one of the easiest things
with respect to this is that it can directly
be placed on top of the Raspberry Pi's gpio pins and thereby making it quite easy
to assemble it as well. So it directly sits on top
of the gpio pins as well. So this is something which makes the entire Raspberry
Pi ecosystem quite easy and effective one other
interesting key point or factor that makes the sensor
quite interesting is that it's gotten
on board joystick as well. So in case if you are a gamer
and you want to try out the games on Raspberry Pi you
can use this choice to catch. As such now another additional
component that most people use with Raspberry Pi is the camera
now in 2013 itself one year after Raspberry Pi
1 was launched. It had introduced
the Raspberry Pi camera along with the firmware
update as well. Now the r a spy cam is
basically an 8 megapixel camera and can directly be connected to the Cs airport present
on the Raspberry Pi itself. Now, it's also very interesting and can record up to
thousand ATP videos as well. So moving forward. Let's also look
at the infrared camera. So this is a very
interesting thing. So this was something that helps you capture
video in infrared mode and is called pine oil. Okay. So again, these components
can be bought online and is one of the most easily by now the final accessori
announce the most interesting and essential accessory
to the Raspberry Pi is the Gert board. Basically it is used
for educational purposes and it helps you expand
the operations done using the general
input output pins here, so, Using this you
can connect to various LEDs which is Analog Devices sensors
and much more. This also helps you connect to Audrey know which is also
another DYI Hardware device which has gained a lot
of popularity but in comparison to Raspberry Pi order, no is still growing
Stitch moving on. Let's begin with
our sensor tutorial and what we'll be doing is we'll be exploring five
different demos to understand how the sense had component
can be used effectively. Now the first thing
that we'll be doing is we'll Implementing
a very simple code and we'll be trying
to display a certain letter on top of the sensor. Okay, so I'm going
to display p&d but the value of that color is going
to be generated by random. Okay, so let me just run
you through the code that's present here. So the first line is
from censored import sense. So what it basically is doing is that it's helping
you import sense at and establish a connection
to the censored component. We're going to use time
for mainly sleeve function. Oh and random for generating
a random integer. Now. What I'm doing is that I'm initializing
the censored which is sense equal to sense at basically initializes it then
what I'm doing is that I'm generating
a random integer between the range of 0
to 255 now for those of you who would have guessed
why I'm doing this. Definitely this is because the RGB value
lies in this range, so it's somewhere
between 0 to 255 then what I'm using is that I'm using
a censored function which is show later and then I'm specifying
which It means to show and I'm going to specify
the color combination. So this is the RGB
values present here. Okay, so it's going to show
the letter e and it's going to create a random input
with respect to that same. Okay, then it's going to sleep for one second again
generate a new random integer and then it is going to show D. Okay. So again, it will sleep
and then it's going to clear. So let me just show you
this to you practically. So now what I'm going to do is that I'm going to connect
to my Raspberry Pi. Using VNC Network. So it's virtual Network, which I'm connecting it to
because it's not quite easy for me to record the video on the Raspberry Pi so I'm going
to use VNC server for that and I'll just give you
a simple idea of how you can connect to VNC server
to your Raspberry Pi as well. Now one essential thing that if you are connecting
to the VNC server, then you both need to be
on the same network. Okay. So what I've done is
I have used a hotspot connected both my Raspberry Pi
and my laptop to the Network so in case you are wondering how to connect to your Raspberry
Pi y remote server, this is how you do it. Just go to your settings
go to preferences. You have Raspberry
Pi configuration here. Now in your interfaces make sure
that VNC is enable. Once you've done this then
VNC server gets enabled and onto your system
just download VNC server and VNC viewer and
you can connect your Raspberry Pi quite easily. Okay. So let me just show you the code which we're going
to be executing now, it's the same that we
have seen earlier. So, let me just make
a small modification to this. So this is awkward again as we have seen we are first
initializing the censored with sense equal to censored then we're generating
a random integer and then we showing a letter e on that so let me
just change this. Let me just set it to 0 and let
me just copy this once more and I'll also show a third
combination here as well. So I'm just generating
a random value between 0 to 255 and I'm displaying
based on that color. Letter e d and f okay, so let me just save this once
so let me show you the output that is going to be
displayed on the sense that as well. So so let me just
execute the program. Okay. Now let me just switch
over to the censored so that you can see the output. It's a ready green d
and a blue f as well. So this is what So based on the inputs that I
have given it has displayed a certain letter on the censored
LED display as well. So let's go back
to our presentation and look at the next program
that is involved here. So the next program basically
is going to display an image. Now what it does here is that it basically is going
to display an image that Gives you a complete idea of the various color
combinations possible on the sense at as well. So what I have basically done is that I have identified the RGB colors red orange
yellow green blue indigo violet and E stands for empty, which is null. Okay and based on this. I've created a
random Matrix here, but each of these values defines
for a pixel of the sensor now as I have mentioned
to you earlier, this is an Eight cross
it led Matrix. So each of these values is for one of the
individual pixels themself and with respect to how they
start the numbering starts from left top position. So accordingly it goes
in a row by row, man. Okay, so let me just show you
the output for this as well. So this is our program
and let me execute this and show it to you. And now let me just switch over
to the Raspberry Pi. To here you can see
the different color combinations that are present
on the Raspberry Pi sensor. The first row was
left completely empty. The next row has
to our values present here. So this is how it
has been defined. If you notice the code here, there are two are following
by three empty spaces. And before that. Also, if you can consider
three empty spaces that is one rule. So every specific
value is pertaining to one specific row as such now
coming back to our presentation. And so the third one
is quite interesting. What we're basically
going to do here is that we're going to rotate a letter based on the different
orientation of the screen. So let me show you how this is done rather
than just tell it to you. So what we basically
going to do here is that we're going
to rotate a letter J on different angles as well. So we've defined
the different angle 0 90 180 270 and these
are the various angles which is going to rotate for. Okay. I'm going to run a for Loop where it rotates
the same on different. Angles as well. So let me just show you
that so just let me add since dot here here. Now. The reason I'm doing
that is mainly to ensure that the sense that is cleared
after the rotation process. So let's execute this. Let me switch over
to the Raspberry Pi and if you see here
A Jay has been displayed and it's rotating. Okay, it has stopped. So let me just rerun
this once again. So what is happening here is that I'm rotating
the letter J by 90 degrees and it's completing to complete
circles on the Raspberry Pi. Okay, so it's a very
simple thing but it's important that you understand
how this is done because what we're
basically doing is that we are setting
the rotation using the sensor. Okay, and then we're sleeping at point zero five
milliseconds as well. Again, you can increase
the delay here. You can change the letter here. You can even change
the angle to understand how it works as well. So this is something that you
should definitely explore about and get a better
understanding with coming down the next and most basic thing that you can do. Using sensor is that you can measure the temperature
pressure and humidity. So we've already
tried to do this in our previous session as well where we've measured
the temperature the pressure and the humidity now here
I'm just rounding it off and based on a specific value. I'm going to ensure whether the display
is going to be red or green as well. So let me come back to my senses
at let me just open the code and we will modify this on the go to get
a better understanding. Let me explain. In the code once again, so first I'm establishing
a connection to the censored then I'm running
an infinite Loop where I'm getting the temperature the
pressure the humidity. I'm rounding it off and
if the temperature is about 36, so this is just a value that I have set then
the background color should be green and the color of the text
should be complete wide. Okay. So this is a simple modification
I have made else in normal situation if it is less than 36 degrees
then what it should be. Is that it should have
a background color of blue and the text color
should be yellow. So this is basically to help me
identify the different ranges. Then I am creating
a message variable which is storing
the following format where temperature is equal to temperature value pressure
is equal to pressure value and humidity is equal
to the humidity values. Finally. I'm going to show
this message on my sense at LED display screen. So let's just see how this works
and execute the program. So let me just execute
the program and let me also switch over
to the Raspberry Pi. So here let's begin
with the execution of the code. So now you can see by default. It is having a blue background and the text is
actually in yellow. Okay, so if you would have
noticed the temperature is about 34 degrees Centigrade. So to this letter or Scroll once more
let's just validate the value of temperature. You can also print
this into the screen. You can also send it
via an API and much more. So if you can see, it's thirty four point
two degree Centigrade now to this what I'm going
to do basically is that I'm just going
to introduce a hot water source. Okay, so I'm just trying
to raise the temperature here. So let me just shift
a little bit and please keep a note
on the temperature value. So the temperature
has started Rising. So it is taking some time. I think the water
has cooled down a bit. All right, so it's started
now the color background color has changed from blue to green and the text color has changed
from yellow to white as one now. These can be used in critical
measure environments as well. So let's say you are
in a situation or an environment where you need to have
a constant Monitor and constantly be aware with respect to the environment
change in example, like space or any other environment
that you're working with. Definitely. These are certain things
that can really be important and helpful as well. So Coming back. Lets us. Look at the last program
of the sensors tutorial, which is basically detecting
the motion of the sense that now what I'm basically
going to do here is that I am going to rotate
my science at and I'm going to identify the value of
pitch roll and yaw on basically, these are the orientation
with respect to the x axis y axis and Z axis and this is just from
the orientation of the sense that as well. Now this is mainly done
using the gyroscope but to get a better
understanding and more precise. Value will also try to include and incorporate
the accelerometer as well. So let me come here. Let me just show
you the program. So this is the program as well. So let me just stop
the execution here. So as I had mentioned this is going to run
infinite number of times because the while statement
here is indefinite. So let me just stop
this and let me just run the orientation program and Let's just see
the default values of pitch yarn and roll first. So now by default if you see pictures
about zero role is about 96 and Ian is about 0 itself. Now, let me just switch
over to the sense at and what I'm going
to do parallely is that let me try radiating
the value of the sensor as well or let me try rotating
the sensor to give you a better understanding of
how it works simultaneously. You can also see the change
in Pitch Ian and road. So if you see Here
I'm slightly lifting it and the value of which
has risen from zero to six role also has changed
Ian also has changed. I'm going to try tilting
it same pitch roll and yaw on also has changed. So I'm just filling it with respect to various accesses
and you can see the change here. You notice the changes I
definitely hope you do. Okay. So now let us actually stop
the execution of this program because this is also
an infinite Loop and to same orientation program. Let us actually
try to incorporate the accelerometer and understand how more precise we can make now instead of just
taking the pitch roll and yaw on I'm here going
to directly take the value of acceleration here. We have taken the value
of orientation and here we're going to take
the Of Aksum duration, okay how basically the first is with respect
to the gravitational force that acts on this. So now let me just show
you the code here. Now again, what you need
to understand the difference between gyro measure
and Axle row measure is just that its senses
both the static as well as Dynamic change
with respect to that but your gyroscope
many measures the rotation of a specific device as such. Okay, so let me just execute
this program as well. so let me just first begin
by clearing the screen and then we'll execute
the code once again, so let's just execute
the acceleration code to get a better understanding
and simultaneously what I'll do is that I'll just show you
how the change with respect to the axis is or how you if you change the Raspberry Pi how the value will also
change so by default if you can see
the value of x is 0 the value of y is also 0
and Z is 1 so now what To do is that let me Begin by slightly
rotating the Raspberry Pi now if I change the Raspberry Pi into a vertical
position like this. Okay, so it's a perfectly
vertical position. But if you notice the value
of x it is become minus 1, okay, why has become 0
and Z has become 0 same time. If I rotate it in such a manner. I'm just keeping it
on different axis the value of x becomes plus 1 instead of
minus 1 same with respect to Y if I change it in such
a horizontal way it is minus 1 and if I change it in such a manner then
it becomes plus one. Okay. If you flip it completely
over then the value of Z becomes minus one as well. So if this is the vertical
position then the value of Z remains one, I hope this is clear
for you guys. So again, these are
all different parameters today what I have done
as part of this is that I've just
help you understand how you can explore
the various parameters and various components that are associated to the Raspberry Pi
censored as well. Now what how you use it and what you use it
for is completely left you these are just some
of the basic things that you can do
using Raspberry Pi the limitation of
Raspberry Pi stops where your imagination stops. Okay, now that we've installed
raspbian on Raspberry Pi it's time we begin with the demo
for this session. Now, what we're trying to do is that we're trying to build
a weather detection system to get the real
time weather analysis when we talked
with real term analysis. It basically is the pressure
temperature and humidity. And what we want to do is that we want to get a live
mobile notification for this. So for that will be first
using censored board which will help us measure
the various metrics like temperature humidity
and pressure. Iran to a Raspberry Pi then
we'll use Raspberry Pi to transfer this data
and display it on the sense that as well. So we'll just be displaying
the temperature as of now on the set. And finally we'll be
using insta push API and Python program through which
we can send this notification across the Internet
to our mobile application. So let's begin
this demo part first. Let me show you
what's and satis and how you can assemble sense
at on Raspberry Pi. Okay. So this is my Raspberry Pi unit and this is That's what
I was referring to. This is sunset boat. Okay. Now if you see this is
a simple additional board that you can add on top
of your Raspberry Pi. Okay. Now it's quite simple, but how it's different is that in comparison
to your Raspberry Pi there is a neat cross 8 digital
LED display present here. So you can use this for various different purposes
as per your requirement. Okay. Now, it's simply sits
on top of the Raspberry Pi. Okay, just make sure
the alignment is correct. And once you've done
with that just push it on top. And the pins would
merge accordingly. Once you do that you
can just place the four screws across the different ends
and tighten it up. And once you're done with this, you can integrate
and use the censored for various different purposes. We will use sensor to measure
the temperature the pressure as well as the humidity
across in this room that is present. Okay, so this has
different sensors present on it and we'll be using them
for today's demo. Okay. So let me just put this up
and once it's all set up. Let's go back to our demo and let's begin
with understanding the programs before we begin the execution. Okay. So moving on now that you've understood
about Raspberry Pi's and Saturn V. We also install
raspbian on our Bye, let's move forward and start with respect
to today's demo part. So again going back
just to remind you. We trying to create
a better control system wherein we try to measure
the temperature the pressure and the humidity
around the sensor as such. Okay. So the first thing we're going
to do is I let me just connect to my Raspberry Pi. Okay. So this is
my Raspberry Pi interface. Now. Let me just show you
the program as well. So I've divided the actual
program into three parts so that it's easier for you to understand how this program
works as such now, the first program that we are going to execute
is the sensing environment. So this is a simple program that's going to sense
the environment variables. That is the temperature
pressure and humidity. And then it's going
to correspondingly display these values on the censored
LED display panel. Now, let me explain this program
to you one by one now firstly we need to import the He boosts
the values on the censored. So for that we're going
to use this line from sense underscore hat import sense
at so this will give me the corresponding values from the sensor
to my Raspberry Pi interface. Then I'm also importing type. Now. The main objective of time is to help me keep
a log of all the values that are measuring. Okay, so that will be
seeing little ahead. After that. I'm going to assign
the censored values to a variable known as sense. Okay, so through this variable. I'm going to correspondingly
relate to all the values coming. From the sense at okay
moving forward we have written a while loop
and this is an infinite while it's going to continuously measure the temperature
pressure and humidity and is going to correspondingly
be displayed on the sense that as well. Okay. So let me help you understand how the temperature humidity
and pressure are measured. Okay. So what you need to do
to measure the temperature is used the function sense dot
get underscore temperature. So this function that is get underscore
temperature will give me the temperature value measured by the sensor
in degree centigrade. It and hear what I'm
basically doing is that I'm just converting
it into fahrenheit. Okay, so the value in degree Centigrade into
1 Point 8 plus 32 is just a Fahrenheit representation
of the current temperature. I'm using a round
function to round of the corresponding value. I don't need any decimal
point values here. I want a whole number. So what I'm doing is that I'm using a round function
and once I have the temperature then I'm going to be storing it
inside a temp variable. Okay. Similarly. We are going to use
sense not get humidity and sense dot get
pressure to get both. The humidity as well as pressure and store it in humidity
and pressure variables are such. So basically I have
all my three variables that are needed now. What I'll do is I'll create
a string called message. Okay into which I'm going to
store the temperature humidity and pressure with
a specific notation. So T is going to be equal to the temperature H is going
to be equal to the humidity and P is going to be equal
to the pressure. Now comes the part where have to display this value
onto my sensor display screen. Okay, so here I'm going
to use the function. Sense dot show
underscore message. So this is basically
a call through which I can send any message to
my sensor display screen. Okay. Now here I am just passing
the message with respect to the temperature humidity and pressure variable and then I'm also
defining a scroll speed. Now. This is something that
you can play around with if you want to have
a fastest-growing you can modify it. I have chosen a 0.08 scroll
speed then the other two factors are the font color
and the background color. Okay. Another text color I
have set as 200 240. Tea and 200 now
if you can guess these are basically the RGB values for
the corresponding text color. So this is something
that I have said as per my requirement you
can play around as well as background color is something that I have set to 0 0 and 0. Okay. So you have complete freedom to modify this as
per your interest. I would definitely recommend
that you try playing around with this to get a better feel
of how it works. Okay. Now what I'm doing is that I'm putting my system
to sleep for four seconds. This is basically to introduce a delay between the
measurement carried out. On the sensor now, if I don't do this since it's a continuously
running infinite loop it's going to keep measuring
at one after the other. So I want a simple delay
between the measurement. So I'm putting the system
to sleep for four seconds. After that. What I'm doing is
that I'm opening a file called weather dot txt now here. I am going to create
the log into this file. I'll take the present time and
then pass the message as well. So these both with respect
to the present time as well as the message are going
to be stored in form of a law. Okay, so in case in future if I want Want to refer
to the measurement, this is something
that will be really helpful. Finally. I'm also going to print
message on my console. Now. This is something just
for our understanding as well. We'll be seeing it
on different places will see it on the console will see it on the sensor as well as we
will be seeing it inside the log file as well. Finally. I'm closing the log and then I'm putting my system
to sleep for 5 seconds. So these sleep delays
are just to ensure that the consistent values
are being measured in regular interval and there is no hindrance
with respect to Okay. I hope you guys
are clear with this because I'm going to go on
to execute this program. So let me open my console. Okay here let me write
the command for executing a Python program
and it's very simple. It's just python followed
by the filename. Okay. Now in our case
it's sensing environment. Okay, so let me execute this. And now if you see
the values are being computed and on the sense that the values are being shown as well so you can see here
the change with respect to that. Okay, firstly it showed me
all the temperature humidity and pressure and now
it's already come on the console as well. Similarly the going to have
a simple delay and then you can Already it showing me
the next values on my sensor and it's going to be
stored in a log file. And then it's going
to come onto my console. So very interesting. Right? So definitely you can play
around with respect to the font. You can play around with respect
to the background color as well as if you are interested, you can try bringing
different temperature items near the sensor to see
the variation with respect to that as well. Let's say if you have
a glass of hot water and you bring it near the sensor the temperature value will
correspondingly change as well. Okay. So this is something that
you can really Experiment on and have a good understanding
of how things work. So let me just close this program and let's
see the log file. Okay. So this is my weather
dot txt file. So I have taken different
temperature measure variables. It's been log from a long time that we've been
trying this out on. So definitely there are
a lot of values here. Okay. So let me just show
you the latest value. So this is a range of values that I had just tried
out some time back. So all the log
all the temperature and pressure log are being
stored here continuously. Okay when you feel comfortable. This is the latest log that we have just created with
respect to the three values. This is the current
status as such. Okay. So again, this is something
that you can experiment on. I am just told it because of
my future reference as well. Okay. So I hope you guys have got
a simple understanding of how you can connect to your sense at measure
the temperature pressure and humidity and also display it on the console as
well as on your sex act. Now coming on to the second part second part
is the slightly Advanced version of the same program here. We're trying to do something
a little more apart from just seeing it on the screen. Okay. Now I want to send an email
with respect to all the logs that I have or let us say
at a regular interval. I want to have an email which notifies me with respect
to the present temperature and the current
pressure and humidity. Now, this is something
that you can definitely use while you're working
on a social experiment as such usually
in the Laboratories. It's important that you
always keep Check on the temperature
pressure and humidity. So this is something
that you can try out. So let me show you the code. Alright, so here I am just including something new as well
apart from our previous code. Okay. Now if you see our previous code
is still present here. I'm taking the temperature. I'm taking the pressure humidity
storing into a message and then I'm also
printing that message but apart from that
I'm using a few libraries which are very essential
for mailing me this detail. Okay. So for that, I'm going to use
first SMTP Library. Okay, because your I'm
wailing Protocols are SMTP based protocols. So these are something
that I will be using in case I need to send a mail
with respect to all the details as such so these
are standard protocols through which I be communicated. Okay. Now coming down with respect to what Mi Ami is
a Miami basically is an extension or multi-purpose
internet mail extension. Okay through which
I can send emails which support text characters non text characters audio
video images and so forth. So sending a mail. Is an essential important
Library as such? Okay moving on forward. Let me explain you the code now here we
are repeating the same thing. We are also importing the time function and then we
are establishing a communication between the sense
at and the program. Okay. So here first you need
to specify the from address that is from Coos mail ID
you're sending this. So I have created
to test Melodies one is a durocher test 11 at gmail.com. So from this address
the mail is going to be sent and the address
which is going to be receiving. This is a deal-breaker. A Raspberry Pi. Okay. So from this email-id
I'm going to send a log with respect to the data that is measured from I
censored Raspberry Pi. Okay, then I'm going to use my Mi Ami multi-part
function from here. I'm going to distribute my mail
into multiple parts. So this is what your Mi Ami multi-part function
will help you to that. So in my message part the from
is going to be the from address that is from whom it
is going the two is going to be the to address
and the subject. I am defining it as temperature
Raspberry Pi so temperature Then I'm measuring
all the temperature as well as the pressure and humidity and storing it
inside a message variable. So here what I'm doing is that I'm basically going
to attach the message in a plane format
to my message variable, which is going
to be the details of the temperature and pressure. Okay. Now, let me come down and help you understand
this part of the code. Now. I'm going to establish
a connection to my server. So I'm going to use
your server variable where and I'm establishing
an SMTP protocol do Google. Okay. That is your or
smtp.gmail.com sensor. This is a standard
mail protocol SMTP for Gmail. It is smtp.gmail.com
other meals you can just replace the corresponding me and then
you have the port number. So again with respect to SMTP. This is a standard Port that you would be using
for communication. That is 25. Okay, then I'm going
to start my server. So that is server dot start TLS. Okay, once I start this then I'm going to specify
the login credentials. This is basically
for authentication Factor. So server DOT log in you have to provide
the Username and password. So my from address is
my username and my editor a kpi is the password
for this email-id. Okay once have specified
this then what I'm doing is that I'm converting
the message as a string. Okay, sometimes what happens is while you are trying
to send this. It may not get sent as such because it has
values also associated with it some directly converting
it completely to a string. Okay, and then storing it
inside a variable known as text. Then I'm calling
the function sin. Email through which I'm
specifying from which address is going to which address
it should go and the text which is part of it. Finally. I'm quitting the connection
between the server. I hope you've understood
till here with respect to the program now, it's time we
execute this program. Okay, so let me call
python email dot p y okay. So now what it has done now,
it's connecting to my server. That's the delay here. Okay, its measured
the temperature its measured the humidity as well as
measured the pressure and then it's going
to send the mail. So let me show you
this meal ideas were so as you can see here. I've already received a mail from anyway car test lab
it's called temp rasp that we had set and it's given me
the corresponding value of temperature humidity
and pressure. Uh sure so this is what
I was referring for so you can do this across to let
me show you my sent box as well. Okay. Now if you see a new Sandbox, I can actually see that this male has been said
so there are two ways for me to verify this as well. So this is something
that is really helpful for me now coming back to my Raspberry Pi let us
now see the final program in which we want
to incorporate this and do a little more. So let me explain you the code
little more better here. There are few things do here. So first you have pikul and Jason being imported
now pikul, basically, Is going to help you establish
a connection between your API. So we're going to use an API through which I'm going
to pass my data. So for that I'm including
bikers here apart from that you have string IO, which is a standard string
input output management package and you have
your our PI dot gpio. So that is Raspberry Pi's
General input output. I am just referring
to it as gpio. Okay, so moving forward
whenever you see gpio it is the general input output
from the Raspberry Pi and then you have the similar
function you have sense. Go hat from censored
and time function and I'm establishing
a connection here as we have done earlier
between my sensor and my application and I am clearing
the screen on sensor. Now. What happens is sometimes if you've executed
a program before and you've stopped the execution
of the program then what happens is the display will still have the remaining
data present on it. It stays stuck on the sensor. So before we move on
to with respect to our program, we will clear the data out. Okay, so that's what
sense dot cleared us now here. Two things that are present
here hot and cold. Okay now why they're
being used I'll come back to that a little ahead. Okay apart from that. There's also a new variable
known as push message. So we'll be talking
about this also when we come down later and you'll understand what
the push message is used for. Now. Let's try to understand the code which will help me
display numbers on my sensor. Now, you might be wondering if I have already displayed
number and data on my sensor. Why do I need a separate
part of the score because here we're trying
to do something different. And from what we
have already done, so here firstly what we're doing
is that we're setting an offset from the left and the top. Let me go back
to my presentation here. Okay. Now here this is the standard program that
we are talking about. But let's see. This is what we need. Okay. So this is what we're going
to explode ahead. Now. This is my actual sense at okay. This is a neat cross
8 LED display function. Okay. So here what I'm going to do
that I'm going to eliminate the First Column and the first two rows
from the top. Okay, so this We will give me a 6 cross 7 Matrix area
remaining with me. Okay, this is where I'm going to do all
my operations from now onwards. So this is something interesting
and I'll help you understand why I am taking this region. Now before we move on let's take
a simple look with respect to the core present here. Okay. Now before I do that, let me try to help
you understand why I have taken a 6 cross 7 Matrix area. Now. This was the remaining area. I have divided this remaining
area into two parts. Okay. Okay, I'm just going
to give a gap of one column between these two and the left hand side is going
to represent the temperature or the value in tens place and the right hand side segment
is going to give me the value in once place. So this basically is
a 3 cross 5 Matrix where and I have 15 positions
to play around with boom. But this number again, this is 15 areas that
I can play around with in a 3 cross Phi Matrix. Now, let's go back
to our code now. Let me help you understand
how it works now. I have basically
10 numbers with me. Okay. So each one is going to have
a specific way of representing in a 3 cross five Matrix. So that's what I
have written here in this number Matrix as such. Okay. So each of the line
is a representation of a corresponding number
in a 3 cross 5 Matrix and this one and zero
basically is a configuration for the LED to be either turned
on or turned off. So normally when we
write 0 the first line will always be full. The first three numbers are one. Okay, then you have 1 0 1. Okay again 1 0 1 1 0 1 and my last three
numbers will be again one so as and when you represent
zero in a normal form, this is how you do that so it may not be so clear
to you right now. Okay. So what I would recommend is before you try to understand how each of the numbers
are displayed set an offset on your sensor there
try playing around with respect to different
values to understand how you can represent each of the numbers once Made
two or three mistakes, then you'll get a clear-cut
understanding of how each of the numbers can be displayed. Now. This is for a 3 cross
Phi Matrix tomorrow. You want to try it across the
across it you can do that as well. So definitely make sure to try out make mistakes only
then will you learn? Okay. So each of the line is
a representation for a digit in a 3 cross Phi Matrix. Now, let's try to understand how we can display
a single digit for that. I'm going to define a function
called show digit, okay. Okay. So this is the function it
takes a value as an input. Okay. It takes a value for x d OK XD is the x-coordinate
for where the digit has to be present yd is
the again the y-coordinate as to where the digit
has to be represented and the last is a simple
RGB value present here. Now first thing
what I'm going to understand is that I'm going to compute
the offset position. Okay. Why do I need to
compute an offset? Because each of the
numbers are present in different rows of the call. So I Need to understand
which is the present number and which value do I
need so for that what I'll do is the incoming
value into 15 will always put me at the start of that room. Let's say the value is 3, okay. So 3 into 15 tells me that I need to
move 45 positions. So as I said
each row has 15 so 15. Okay, then I have 30
in the next row. Okay, and once I'm done with
the 45th transfer then I'm here. This is the standard
representation for three. 3 so I have this complete row
at my disposal presently. I am pointing to the start
of three value. Okay. So this is something
important to remember. So this is why we
are Computing an offset by multiplying the value in 250. So I have the complete
access to this row. Okay now we'll try to execute a for Loop through which
I'll place these values into the different parts
of the LED display. Okay. Now see I have
three cross 15 area. So I need to identify
in which area should which Look up that is
which led should be turned on. I have three rows and 5 columns. So I need to identify this. Let's go back. Now. The first thing that I'm going to do is
that I'm going to compute x t OK XD basically is going
to be the value of p modulo 3. Okay. This basically is going
to give me value of 0 1 or 2. Okay. So if you go back
to our presentation, it's either going to start
from this 0th position or it is going to be in the first position or in
the second position. Okay, similarly in
the left hand side. Also, it could be
this 0th position first position or second position. So basically we are
dividing this that's what you need to understand. Okay coming back. Let me just remove this. Okay. Similarly. I'm going to compute my y of T by have to basically
is to help me understand which value should I take
from this complete row, so this will basically start
from zero and go all the way till 14 as such. Okay, so it will basically
represent which column value should I take So I
hope you understand from this which column value
should I keep taking for my next value. Once I have this then what I'll do is I
am going to call the sense dot set pixel Rumba
how we had called sense - roh message so there and we had sent
a complete message. Now, what we're doing is
that we just sent to each of the pixels
individually as such. Okay, so we're going
to use the x coordinate. Okay, which is going
to be my x t plus XD. Okay. Remember this? Okay. We'll Define X. Still it'll add but just
keep this in mind XT plus XD is going to give
me the x coordinate y t + yd is going to give
me the y coordinate and then basically
my RGB number comes so here I have just done
a multiplication of P into R. So whichever I want I'll
just convert it accordingly. Okay. Now comes the choice of when you have to display
two digit number. So instead of going back
to the same function. F defined a new function, which is called show number now show number is
The original function that will be caught because the temperature measured is most often
a two digit number. So this is where the actual code
was going to start from. Okay. So number takes four values. It takes a variable called while which is
the present temperature. I have my RGB value as well as
I'm going to begin the function here firstly I'll begin by taking the absolute value
of the given input. Okay, when I say absolute value, I am basically trying
to identify the number present in the tens place. He's okay. So this is what absolute value will be whatever the digit
is present in the tens place. I'm going to store it
in my abs underscore value so that I'm just
providing it here to the tens place then
let me also identify what is in units place and this
is simple mathematical operation where I divide
the number by 10, I'm going to get a remainder and that's what I want. So I'm going to use
a modular douche. So my absolute value divided by 10 is going to give me
the unit's place as well. Now, what I'll do is I'll run. If condition so here
what I'm going to do if the value is greater than 9 then I'm going to call
the show digit function which I had defined earlier and I'm going to pass
the tents value. Okay, the value present
in the tens place. So and then I'm going
to also pass the XD n yd. In this case x d + yd
is going to be the offset left and offset top why let
me help you understand here. My offset top is going to be
with respect to this position. It's going to start from here. Okay, so it will point
to the third row. Okay my offset. Left will point
to Second column here. So I know that this is
where my tens place has to come. So that's why
I'm passing it here. Okay. I also am going to pass
the standard RGB value that I have got input here then
for a second time. I'm going to call the show
digit function where and I will pass the unit's place, but with an offset
of additional four spaces. Okay. So this coming back
to our segmentation. This is going to be
four positions from our left. So offset left is going to give
me the first place but offset, Plus four is going
to be the start place where I have defined
for my units place when I say from top it's worth
going to be the same because I'm not separating with
respect to different levels. So I'm just going to change
with respect to my offset left. Okay. So that's the only difference
present here now coming down. We are Computing the basic
temperature the humidity and pressure and then storing it into a message string
now comes the place where I need to do
my actual implementation now if you remember
our problem statement, We have to send
the pressure temperature and humidity as a notification
on our mobile device. Okay, so for that I'm going
to use in stop push which is an application
through which I can send notifications across my mobile. So my python application that is the final application
will communicate to install push and that will
in turn communicate to my mobile application that on which I
have configured it. Okay. So the notification
will be received on my mobile application
now coming down. Me remind you again
why we are using this in a real world scenario. This is something
that is being used by NASA. Okay, they're they're using
it to identify if the exterior temperature is
falling between a specific range or even let's say if there's a change
in the pressure inside the cabin or if the humidity is too high, so all this can be measured and you can get notified
with respect to any change that you want. Okay now coming back
to our program. This is why I have defined
a range of cold and hot If it falls below 37 degrees, I want a notification
if it goes beyond 40, I want a notification. So this is my range by value. It should either be
only 38 or 39. Okay. So this is why I am using
cold and hot variables now. Let me help you understand how the in stock push
application is working for that. Let me open my browser
and show you how you can configure in stock push for
your Raspberry Pi application. Okay. So this is
my insta push homepage. Okay now first. Seeing what you need to do is that you need to begin
by creating a user ID credential here and it is very important. When you logging
into your application, you use the same credentials. So make sure you
do not lose this because it's very important only if you use the same
credential will you get this notification so remember that so you have an option
to sign up for free as well? Okay. All you need to do is
provide your username provide your email address
and then specify a password now, let me just log in. Okay, so I'm just logged in. Okay now here you can see
there's already a timeline because I have already
configured this application. Let me help you understand how you can do this now
in my application. I already have a temp
notify application. Okay, which will be using as
part of our understanding. Okay, firstly let me click
on ADD application now once I do this I need
to specify a name. So let me just call it sample. Once I click on Sample once I press add
application sample gets created now it is asking me
to add event here. So click on add event. So let me call this temp rasp. Okay, and then tracker
would be message. Make sure you create a tab here
after you specify whatever it is that you specify and here push
message will be message. So make sure it's present
inside curly braces. What are we are specifying here. Once this is done just click on add event and this event
is going to get added. Okay, so the event
has been added now it's time. I helped you understand
why I had specified message there so rumor let me go back. To my code. Remember we had defined
something known as push message. So this is why I
have defined message here. So this is what is going to be sent between
my final program application and my insta push application. So this is why we had defined
this variable earlier as well. Okay now coming down
with respect to this. Let me go back to my insta push
and let me help you understand. What all do you need
before you move ahead? Okay, just go to the basic info Tab and here there are
two things that you need. Okay first. First you need
your application ID and then you need
the application secret. So this is very similar to how
you have a user ID and password. So this is what you're going
to use in case you're working with an application. Okay. So make sure you note
this application ID and the corresponding
application password. Now if you come back
to my program you can already see I have mentioned
the application ID as well as the application secret and here the push event is
to my temp notify. So this was the previous event
that was already set. So I will be reviewing. Bring to that as well. Now, what you need to do is that you need to use Curl
to post whatever data that you have
to the insta push API. So for that you're going
to use biker dot curl function. Okay, and this in turn
will help you communicate between your application
and insta Porsche. Okay. So all the communication
is going to be through a variable called
C. Now C dot set up. Okay here what I
am going to specify is that I'm going to specify
the URL first, okay. The URL is going to be HTTP colon slash slash
a pi dot insta pushed IM / version 1 / post. Okay. There's also another
alternative URL which is HTTP colon slash slash
a pi dot insta pushed IM / post itself now in case the other is not working you
can use this or vice versa. Okay, moving forward. I need to set up
the customer headers basically for my authentication
and content type specification. Okay. So for that you're going to use
the command C dot set of forward by C dot HTTP header. Then I'm going to specify x - install push -
a pi D plus my app ID, which I have specified earlier. So this will help
me identify to which up. This has to be pushed. Okay. The next is going to be x -
insta push - app secret. So basically the app secret
your your password in this case. Okay, then comes
the content type. That is what What type
of content I go to pass so this is going to be
application dot Json. So I'm basically going
to pass a Json file to my API. Okay. So I hope you guys
are clear till here. Okay. Now what I'm going to do is that I'm going to
define the function through which the message
is going to be passed. Okay. So here what I'm going
to do first is that I'm going to use
a dictionary structure for passing the data as Json
in my post function. So Jason - field is going to
create a dictionary and in through that
I'm going to send this data. Okay. Now what I have to do is that I have to specify
the field values. So first comes the event, okay, which is basically
what event I need to relate to then comes my tracker. I'm not specified any tracker
as such then comes my message where and I'm going to pass
the push message push message basically is the
temperature the pressure and the humidity as such now if you want you can print this but I am not going
to print it here. Finally. I'm going to use Json dot dump
to make a Json file and going to assign it
to post Fields variable. Okay. So post field is the variable which I am going
to pass to my APA. Okay next to send your Json file with post use the following
command C dot set-top. Okay inside which you're going
to specify CDOT post field and you're going
to pass this post field, which is the Json file as such then if you wish
to capture the response, then you can capture
it inside a buffer. So for that I am also going to
use a buffer to write operation and this can Done
with the following command C dot set up inside which I'm going
to write C dot right function and buffer, right? That is the right back
has to be into my buffer. Okay. Now if you want to be updated whether the post
has been sent then you can use the following command that is C dot set top followed
by C of C were Burrows. So this will make sure that I am getting posted
on my terminal with respect to the communication finally
comes our operation of checking the temperature. This is very similar
to what we have. I've done earlier where we are running
an infinite Loop. I'm measuring my temperature
in degree Centigrade. I'm measuring my humidity
measuring my pressure storing this value
inside a message variable and then putting it
to sleep simultaneously. I am also writing
into a log file and this is going
to be stored with respect to the present time. Okay, so till here I
hope you guys are clear now comes the part where I
have identified my temperature. I'm just making sure
it's an integer and then I'm going to call
my show number function. But if we had defined a function based on which it
is going to be displayed on my Raspberry Pi so this
is a call to that function. Okay, once I
have done with this, I'm going to store it into temp
1 then I will make sure that I am writing it
into my file and then comes the interesting part if my present temperature is
either greater than or equal to the hot message or if it is lesser than or equal
to a cold temperature that we have defined rumor
the two variables. We had defined earlier. This is where they
are being used. Used. Okay, the push message
will start with either if it's hot or if it's code following that would be
the present values. Okay, the pressure the
temperature and the humidity. Okay, then what I'm doing is that I'm going to pass this push
message to my P function that I have just defined. So basically it will first
convert the push message into a Json data and then it will pass it
through the communication. So this is
what is happening here. Okay. Same thing will happen in case
it is the temperature Falls below the cold rain. Guess what? Okay, then I'm going
to create a Json file and then call CDOT perform. So this basically will initiate
the curl operation to start finally to capture
the response from the server. You can actually
use the buffer rage. Okay. So with this we have come
to a conclusion. Let me just show you
the final part of this. Okay, so make sure that you're closing
the communication through curl. So you see door closed. Okay. This will ensure that the communication done
through curl is closed as well as you ensure that the General purpose
input output is cleaned up so that new values
are being taken. So now let's execute
the final program. That's python final
program dot P by okay. Now as you see on
the Raspberry Pi it showing me that it's presently
35 degrees Centigrade, okay. Okay, I've started getting
notifications on my phone. Let me just share
this phone screen with you. Okay, and you can see there's
a variation with respect to the temperature shown
on the Raspberry Pi as well. Okay, let me show you
first my phone screen. Now. If you see here it's actually showing multiple
notifications being generated. Okay, it's showing me
that it's cold presently. And here's the complete list of
the temperature notifications. So I'm getting a constant
notification with respect to the change in temperature. So you can see it's continuously
being Generated here as well. So this is something
that is really helpful when you're working
around with respect to a real-time scenario. So like I said NASA
extensively uses this for their own purposes of measuring the
temperature pressure both in their scientific research
area as well as the shuttles and space stations
that the user. Okay. So with this we come
to a conclusion of our demo session Raspberry Pi
has become the heart. Of The Internet of Things domain and camera is one
of the most interesting modules that you can add onto this
interesting Hardware as well. Now, let's begin by exploring
the pie camera for that matter. Now the pie camera was one of the best additions
to the Raspberry Pi component because this module has helped
the users to use the perspective of camera into one
of the smallest and the most effective combination
of the internet of things. Now the main objective of the by camera module is
to help you capture images. It's capture videos time-lapses
even slow motions as well. Now apart from this
the pie camera also helps you capture videos in full HD
at 30 frames HD at 60 frames and even Vijay videos
at 90 frames as well. Now there are two versions
of my camera that are out there. Now the latest version being
the camera module version 2, which was introduced
in 2060 now, although there's not much
of a difference between both these cameras. There has been a slight increase in the resolution wherein
you had five megapixels for the first module you have a It's not and even the sensor
on board has been swapped out where in there was omnivision
for the first version now, it has been replaced by
a Sony sensor as well. Now talking about the by camera. You need to be quite careful while you're working out
with the by camera. So let me help you understand how you can set up by camera
with your Raspberry Pi as well. Now, there are five different
stages on setting up by camera will be looking at each one
of them one by one first. Let me help you understand
how to connect your pie camera to the Raspberry Pi now
this Is very important because we've used
three camera modules and we managed to burn
out two of them. So please be careful because this is
a very delicate ribbon and this is also
a very delicate module as such so let me
help you understand how you can attach the CSI cable
to the Raspberry Pi and then I'll also
help you understand. Once you've done this how to enable the camera as part
of your Raspberry Pi as well. So this is
my Raspberry Pi camera, and this is the version
2 of the camera now if you notice this is the camera
that you can see and this is something
that's really difficult. It gets a piece be careful. Now one thing that I really want you to notice
here is the connector here. So this is called The Sunny Side
Up and this is called the silver and now you need
to be really careful when you're plugging this so, let me just bring in
my Raspberry Pi here as well. So here we have
our Raspberry Pi 3, and I'll just show you
how to connect this. There are mainly two precautions
that you really need to consider while you're working
with the Raspberry Pi camera. One thing is just please ensure that the Raspberry Pi is
Switched Off or disconnected before you connect
the Raspberry Pi. Camera as well. The this actually leads to a high chance
of blowing up your camera. We've already successfully
managed to do that once so, please ensure firstly that the Raspberry Pi is
Switched Off and also that the connection is correct. If you notice the silver
side present here, this should be facing
the HDMI port that is present here. Now in order to connect this
let me just slightly push it off and this slot needs
to be slightly raised up. You insert it like this. Okay. Please be very careful. Once you've successfully
inserted it then just push down the hinge as well. This will lock it. So make sure the silver side is
towards the HDMI and the Sunnyside is
to your 3.5. Mm Jack side as well. Now. It's recommended that you get
a case wherein you can also support the camera because it's slightly
challenging because there are rubber in itself is delicate. So when you get a case, which has the camera support,
it will be really useful to you as well. So it is a very delicate items. So please be careful
while you're working. Thing with the Raspberry
Pi camera module moving on. Let's look at how to go
into the interface and enable the Raspberry Pi camera module
for your Raspberry Pi as well. So now that we
have successfully connected to our Raspberry Pi
let me show you how to connect or enable
the Raspberry Pi camera on your Raspberry Pi as well. Now, what I have done is that I have established
a remote connection between my Raspberry Pi and my system using
VNC viewer as well. So this makes it quite easy
for me to show you what I'm doing
on my Raspberry Pi. So once you have successfully
connected and turn down your Raspberry Pi you
can go into the menu and then go to the preference
section here going. To the Raspberry
Pi configuration. And what you need to do is go
into the interface section here. Once you're here make sure that you have enabled the camera
option as well by default. This would be disabled. Now. I have any able most of these
for different purposes so you can just enable
the camera option as well. Once you're done with this then
we'll go on to the next stage, which is understanding how you can capture images
from the Raspberry Pi camera and how to capture video
from Raspberry Pi camera as well now rasp eyes still
is actually a shell command. Which lets you Is
the camera module and capture an image now, the command for capturing
an image is rasp. I still - oh and name of the image that
you want to store it as now - oh is basically
to open the camera and capture the image
and at the same time if you want to rotate the image
vertically or horizontally, you just need to pass
the parameter - VF or - Hatcher now both of this
will either flip it vertically or flip it horizontally
and when you pass both of them, it will completely
flip it around as well now apart from this if you wish to capture a video from the Raspberry Pi
you can use the command r a spy with doing
this will help you again capture the video
from the camera module. Now, this is something
that gets stored by default as a h.264 video as well. If you wish to convert this you
can use the following command, which is sudo
apt-get install - why g-pack now? This video will actually
help you install MP4 box onto your machine and help you convert
your h.264 videos into MP4 format as well. Now once you have it
in MP4 format then Quite easy to play this video
and check it out as well. Now again, if you want to run it
for a fixed duration, then what you can do is
that you can also pass the duration as part
of the parameters while you're specifying
a spy with as well. Now one thing that you
need to keep in mind while you're doing this is that the timeframe that you're passing
here is in milliseconds. So please be careful
while you're doing this now, although I am specifying 10,000. It's only for 10 seconds that the video is going
to be captured as well. Now, let me help you understand
how you can do all this. By using a python function
rather than a shell command as well for that. What you need to do is that you need to install
a package called by camera. Once you've done this
then you can easily access or you can easily perform
all the operations as part of your Python program using
the pike Ram module as well. Now, what you need to do is that first you need to update
all the packages that you want. Then you need to install the python by camera
package that's present. Now if you have two versions
of python on your machine that I would recommend
that you run both these commands that is python - by camera and by In 3 - bye camera this
in turn will ensure that you have the 2.7 and version as well as
the three-point X version of Pi camera present
on your machine as well. You can also install Pi camera
using pip install as well. So the command is pseudo pip
install Pi camera as such moving on let's talk about how you
can actually create a program or how you can use
a Python program to record the videos
for the same 10 seconds. Now earlier. What I had done is that I had just mentioned
that run for 10 seconds using the Shell command r a spy
with but Technically speaking. This does not really account
for all the challenges that may come across while you're actually
capturing a camera now that is where you use a Python program
to help you ensure that even if an exception is come across then
you can handle it and due to that will
be actually using the with statement here firstly what I would like to do. I would like to import
the time package of now importing the time package
also helps me access to the pi camera module and the by camera function
will in itself connect to the pie camera module
and the data that comes With them
not when I'm using with function here this in turn will actually help me
in handling all the exceptions and encapsulating
the proper tasks and cleaning up
the tasks as well. So basically becomes
a best practice while you're using the by camera or while writing any codes
for that matter. So most of us programmers
would like to consider all the scenarios
and take care of it. And in this case python
has given as the with statement to it take care of it as well. Now if I use the command
camera dot start with you, what is going to happen is that it's going to give me
a Review of what my camera is C and I call a sleeve
for 10 seconds this in turn will ensure
that for the next 10 seconds. The camera preview is
on once that's done. I'm going to stop the preview. So what I have here is a preview of what my camera is seeing
for 10 seconds as well. But this is not
where it restricts. There are various other commands
that my pie camera brings in as well talking
about these commands when I use the command
camera dot capture. It captures an image
and Camera dot start recording and stop recording
will in turn help me. Did a video as well if I go back
to my previous program here here I can also
write camera to capture in order to capture the image
between the 10 seconds or what I can also do is that I can also start recording
at the same time as well. So I started recording Let
it wait for 10 seconds, and then I stop
the recording as well. So these are different commands that you usually use
together apart from this if you want to capture a stream, then you can use camera to capture my streams followed
by what type of stream that you want to store
it as well now if you wish To stream the video that is coming from your
Raspberry Pi then you can use the command camera dot captures
followed by my stream and the type of stream that you wish to store. Now again, this also can be recorded wearing you use
camera dot start record. It captures the stream
and you specify the quantization that is needed to this as well. Again, if you wish to play
with some of these properties of the camera as well, you can play around
with the brightness. You can change the resolution and you can even add a weight
with respect to it. Similarly as you have a timer function in
your standard cameras the same. Be used here as well. So let's say I want to start
the recording after 5 seconds. I can do that as well. Now with this, let's come to the demo
of this session. Now, this is a slightly
interesting demo. I'm quite sure most
of you are looking forward to this as well. Then we'll try to
help you understand how you can actually set
up a security camera at your home as well. Now talking about the overall
system here now here, what I would like to do is that I would like to capture
the image of anyone who comes to my door. Once that's done I would check if this person is a recognized
person or is it someone that already has
access to the house? If not, I can always
prompt my user and And give access to it. If not, I can deny access to it. Now to those of you actually see
in our iot projects video. You would have noticed
that in Mark Zuckerberg section. He had actually
allowed his parents. Now, I do not want
to manually have to do this when I just automatically allow
people to do this as well. That's a smart system
or that's an enhanced system that we really would love
to achieve as well. So we would not be
completely able to achieve it but we will take
our tiny steps towards it and we can all enhance it as
per our requirements as well. So firstly what
We're going to do is that we're going
to detect the face. We're going to gather this data. We're going to compare it and then we're going to start
recognizing data as well. Now again, this is
a learning process that takes place. This is something
that involves a lot of understanding and this
slightly gets interesting for most of you as well. Now when we talk
about face detection now, the first stage of face detection is
Verena program actually decide whether an image has a face
or it does not and this stage in itself is called
a classifier as well. In order to achieve this opencv
actually provides us with to pre trained and ready to use
face detection classifiers, which are part of it. We have the half classifier and we have the LBP
classifier as well. Now one thing that you
need to notice or one thing that you need to keep in mind while you're working
with this is that the classifiers
themselves process the image while they are in grayscale
as well because it gives the machine quite Clarity on whether to classify it
or not as such now when I talk about each
of the classifiers whether I talk about
the heart classifier or the LBP classifier, Each has its pros and cons. So let's actually try
to understand this. Now the concept of phase
detection is not something that's quite alien to most
of you and I'm quite sure you've all seen it be it
in our Facebook applications Snapchat. Some of us have also seen it
on the ATM machines as well. We're in Phase
detection is incorporated. And if you wish the most
simplest methodology or if you want to see
the most common example that's out there
for face detection or face recognition. You can see it in most of the phones today
as well again here. What happens is
that The two algorithms that are most widely used are
the higher classifier and the LBP classifier as well. So again, both of these classify are mostly processing
the images in grayscale because we don't actually need the color related
information to decide if there is a face
or not now again, there are going to be
two classifications present here the computer program
which actually decide whether there is
a positive aspect where in the phase is present or the negative aspect where
the face is not present or so. So this is how you
decide or this is what you Call a classifier
for that matter now in opencv as we have seen. There are two
different classifiers, which is the higher classifier
and the LBP classifier as well. Now again, when you talk
about the higher class where it is actually
a machine learning approach that we take a and this was created by
Paul y Le and Michael Jones wherein they are trained
many positive images and negative images
without faces as well. Now this intern actually
started helping them understand that extracting feature
takes quite some time and if you consider
both What segments when you consider higher and LBP had actually takes
up slightly more time. But at the same time it gives you a slightly more
precise solution as well. Now what it actually does
is it actually considers the adjacent pixels
to the region where the face is protected and then it actually
sums up the intensity of these pixels in each
of these regions as well. Once it's done. It actually calculates the
difference between these sums and then it uses
them to differentiate between the section and the subsection
of the Ages as well now when it comes to the LBP
section of this here, it creates a 3 cross 3 window
at the time of image as well. Then what happens is that it starts comparing
from the Centre pixel with all the other pixels which are greater
or equal to the value which is set as 1
or 0 as well then again, it goes on to read
the binary value in a counterclockwise order
and converts it to decimal which are Center values
of the pixel as well. Now what happens is that this process is completely
repeated one after the other and you An easier or a smarter
solution to this as well. Now again, as we have discussed both
of them have an advantage and a disadvantage as we are talking
about the half classifier there. It has a very high accuracy
when it comes to detection of face and there's quite
minimalistic false positive. But the challenge is that the computational time
that it takes is quite slow. And again, the training time that it takes is quite long as
well and even in dark segments, it's not quite accurate as well
because it fires High Clarity with respect to the images
or the scanning area that it takes in the picture. But at the same time when you look at LBP
classifier its computation and training time is actually
quite fast and shorter when you compare
to the hard classified as well now it is quite robust when to any elevation changes that is when there's
a change in lighting. It's still comfortable
and capable of handling this but at the same time
it's not quite accurate or it's not on the same level
at the same time has a high chance of false
positive fatigue as well now. Move on forehead. Let me just show you about
these packages slightly better. Now. If you come into
your opencv package now, I would recommend that you download the opencv
library from GitHub itself. And if you come
inside the segment and you go inside data what you can see here are
the different hard classifiers and LBP Castle path. Now, if you go inside
the hair Cascade section here, you can find different
XML files present here, which will help you understand or help you recognize
different aspects of a body it can start from I you
can start go to I classes And then sue forward as but now
what will mainly be using is that we'll be using
the frontal face segment here. Okay. And again, these are XML files. So even if you open them
and if you see them you would not perfectly
get a clear-cut idea but this is something that we have to use as part
of our program as well. Meanwhile while this is opening. Let me go back. Let me show you
the lpp segment as well. Okay. Now what you see here
are different values that are associated in these are different values
to help you identify or help you understand how the face recognition
should take place as such. Okay. No again. It's a very huge document. So what they're trying to do is that it again going to try
to restructure it again. I think you've got
a simple idea at this point. So let me just close
this and let me go back now if you check the LBP Cascade segment as well here
you have different segments. So again, what we'll be using is that we'll be using the frontal
face segment as such. So what I've done is that I've just created
a separate folder where I've kept my program and these two opencv files
for different algorithms as well now coming back we've done with our
detecting the face. Let's talk about Using our faces as well now till here
what I have done is that I have detected my face, but I need to start
recognizing this face as well. So face recognition mainly can be classified into
three different steps first. I've actually dealing with the data Gathering stage
now in the stage, what you're mostly
going to do is that you're going to gather
any relevant information with respect to that face. So usually this is done
when you have multiple photos of a person and then
you start identifying or you start training
your machine on this face and then it starts recognizing
the Face every time that it comes up with an image
of the same as well. Now there are
three different functions that opencv provides us to do this and let me give you
a brief idea on each of them so that you know, which to use well now first
comes the Eugene faces now Eugene freezes actually
takes all the images at once looks for all the important and useful
components in these phases which are known
as principal features and then passes it
on now one key factor that makes or one key challenge
with respect to the Eugene. Phase function is that it requires a good
amount of Illumination. But at the same time when you have your fish your
face here individually images or individually features
are extracted rather than extracting one by one and this in turn actually does not have a challenge
with the illumination segment as such third is your lbph, which is your low
binary pattern histogram and this is very similar
to your LBP as but but what happens here is that you create a binary Matrix
and convert into decimal factor and then you actually Not
a histogram with respect to it. So what happens here is
actually a comparison with respect to the histograms
that are present. So each phase will
have a correspondingly different histogram and I just compare
these histograms to in order to identify which phase
it is belonging to now. Let me come back to my place where you buy and
before I go forward as we said this segment
actually deals with two parts first is actually
capturing the detection and the second is
with the recognition part. So let me first show you
how you can detect it now. We've already discussed above. About detection or how to
capture any intrusion detection in our Raspberry Pi 3 tutorial where we actually capture
someone opening the door and coming in as well. Once I've captured this image. Then I need to start
recognizing now first. Let me show you
this program here. If you want more details on how it works out then
you can definitely check out our Raspberry Pi
3 tutorial as well. Now if I use the following code,
let me just expand this a bit so that I think you
may not be visible to you. We just increase the font size. Yeah, I know. I think you can all see
my code as well. Now again, we're using
different packages here you using user Library CV
to the time and numpy as well. Now, I'm first going to set the first frame
then I'm going to decide whether there's any
detection as well. Okay. Now for this as we have mentioned
in our Raspberry Pi 3 tutorial, you can use the IP
camera application which is an Android application and you get an IP address
as well or a URL which have specified here. Now, let me just run
this and show you what I'm seeing at this moment. Now what it's going
to do also is that it's going
to capture the last image that the camera sees as well. So any change
with respect to it? It's going to overwrite until I quit or I exit
from the program as such so let me just run this. I just zoom in a bit. So let me go back to my desktop and I'll run
the camera application. Now what it's doing is that it's connecting
to my phone camera and it's going to create or show
you what I'm seeing or what. My phone is seeing
at the moment. So what you see in is
my Raspberry Pi set up here as you can see, it's set up on a curve
because it's slightly tends to get heated up, but you can see it's
a separate device as well. So this is my first frame
and what I'm going to do is that I'm going to put my hand
in front of it and I'm going to make this as my last frame
as well while this remains. I'm going to press
Q to quit from this. And it also has created an image called test
dot jpg here as well. Let me just open this and if you see it's my hand
that has been captured. So what happens here is that it captures the last
in frame as well. So if you want to get
a better understanding of which image it captures
will rerun this but before that, let me just remove this file. Okay. So let me just send
it to the waste bin. And then what we'll do is we
just re execute this program. I'll start by keeping my hand in
the front of the camera itself. So this is my first frame, okay. And any change to this
let me I'll just move it off completely. So this is going
to be my last frame and I press Q to quit from this and let's
just recheck the image as well. So you can see it
has captured the last frame as such with respect to this. So it's a very interesting thing
for capturing this and now let's actually come
now into our program which talks about
face recognition and face detection as well. So this is the program that will help me detect
the face train my application and also understand which face
belongs to whom as well. So I have imported my opencv
to because that is where my face detection
and face recognition. Algorithms are present. I'm going to use
the OS module here as well because I need access to the drives our the path where these images are going
to be stored as well and numpy is one of the very
Key Programs of is one of the very key modules, which is required for any face
detection recognition as well. Because the information that you're going to be using
would be stored as a numpy array for that matter. Now again, what I'm going to do is
that I'm going to actually identify the face
of three people. Okay. I'm going to identify
whether the image that I'm passing is of Ronaldo
is of messy or is off Robin. Okay. Our if it does not fall
into either of these three, it's going to give
it as null as well. Now our first program
or the first function that we're going to be
working is with respect to detecting the fixed. Now, as I said, all the images are going
to be converted into grayscale. So that's what I'm
going to be using. C V2 dot CVT color
to grayscale conversion. Once this is done. I'm going to create a face
Cascade for that matter. Now C V2 dot Cascade classifier
actually specifies which kind of classifier to be used. Now for this program. We're going to be using
the LBP classifier. Okay. And as you see it's
for the frontal face as well. We've seen there are different
aspects but we're going to be using LBP frontal
face classifier as such now this phase is going to store
the list of all the faces that I'm going to be detecting
or I'm going to be identifying. Okay. So for this what I'm
going to do is is that I'm going to specify
face underscore casket and then I'm going
to ask it to detect in multi scale parameter. So I'm going to say it has
to identify in grayscale. The scaling factor of the image is set as 1 Point
2 and the minimum number of pixels that it
should consider is 5 okay. Now if no faces are detected
from the images that I pass then it has
to pass the parameter as none. But if it is detected then I
want all the four coordinates with respect to
the X and y-axis, which I'm going to be storing
in XY W and Okay, X Y width and height
with respect to smarter or I'm going to specify
which part of the image does the face actually contain. So I'm going to specify with
respect to the four coordinates that I'm going to be passing. Once this is done. Then I'm going to define
the next function which is to prepare my data now as we've seen first comes
detection then comes preparing when it comes to it. I need to specify the path from where it has to pick up
the images for that matter. So let me come back. Let me show you
where the images are. Isn't now when you see
inside training data folder. Okay, I have actually
three folders present here, which is my S 1 S 2 and S 3
if I go inside as one you can see there is
a Ronaldo's photos. If I go back
into S 2 you can see that is Messi's photos. And if you see inside S3, this Robbins photos
present here now. Let me come back
to the core now here. What I'm doing is
that I'm passing the path then I'm calling the phase function
as well as the labels. Now again, the phase is going
to hold the sub phases of all the subjects
and labels actually. Help me identify which
it belongs to as well. Now what it's going to do is that it's going to run a for
Loop wherein it goes through. Each of the images
is present in the path. Now again, what we have decided
or what we have started with the naming convention
is with s 1 2 3. So what it's going to do is
that it's going to check if the folder name starts with s because usually there are
system photos separately. So it is going to check
if the folder name starts with s if it does not then it's going
to go into the next Loop. But if it does it will check
that folder as well. Then what is going to do is that it's going to create
the label for it or it's gonna identify
which folder I'm presented. If you actually go back inside
back to my training data, you can see I have specified
1 2 3 after this so it's just to help me identify which label
it belongs to as well. You can use any different
naming conventions that you want as such then
what I'm going to do is that I'm going to specify
where the folder or where the images is present, which is actually
one level inside this so I'm going to go
into the next directory. And then what I'm going to do is that I'm going to run a for
Loop for all the images. Are present in this if an image or a file
for that matter starts with DOT because system files
usually start with DOT. So those are going to be ignored but others are going
to be recognized using my opencv where and I'm going
to read the image first. I'm going to show you a message
that says training on the data. I'm going to resize
the image to a size that is comfortable for me, which is 400 pixel
to find a picture. Okay, and then I'm going to wait
four hundred milliseconds for that matter. Now what you can see here is that I'm calling
to detect phase function which is present earlier. Yeah, so this again is a call
back to the previous function where it goes up and it actually
calls the previous function that we have created
to detect the face as well. Once that is done. It's also going to revert with me the position
as well as the face or the ID which it
belongs to as well. Now if a phase is present
then I'm going to append it to my face list and I'm also going to add
a label to this face as well. Once this is done. I'm going to
destroy these windows because again opencv
creates a lot of windows, so I'm going to ensure that it's destroyed
by doing it twice as well. This is done. I'm going to return the This which I have and the labels
correspondingly as well. Now what you need
to understand is that here the faces here are
of different players as well. So each of the phases present in each of these images
is going to be segregated. It's going to store
this into a location and it's going to start
raining on them as well. Then what I'm going to do is that I'm going to prepare
the training data. Now this again is something that is becoming a recursive
function for that matter. It's going to be
calling itself again where it goes back it
goes to one level up and it's going to
keep calling itself until I have trained
on all the folders which are present and all the images present
inside that folder as well. So once this is done what happens is
that it prints me that data has been prepared. And then it also tells me
the total number of faces and the total number
of labels also present here as such finally
what I'm going to do is that I'm going to create
our LVH be faced a cognizer. So for this I'm going
to call the lbs be faced a cognizer function. So if you remember
we had discussed about three different
phase secondary functions, we're going to be using
the LBH be faced a cognizer function figure. The function has been defined then I'm going
to pass these faces so that my machine can learn and train itself to identify
these phases in future. Now, what I'm going to do is that I'm going to just create
a rectangle across their faces. So that becomes
a differentiating factor for them or helps them separate
where the faces with respect where and once
this is defined or once I have created a rectangle
across their face again, these XY heads that are the different
coordinates of the face, and I'm just going to draw
a rectangle across them. And again, I'm going
to specify which color so this is your RGB value. So it's going to be
a green color box that's going to be defined
and I'm also going to print the name
of the person present here. So what it's going to do is that it's going to identify
which person it belongs to and it's going to print
it in hen seaplane. Okay. And what is going to do is that it's going to be printed
again in green color as well. Now what I comes or now the most
important function comes which is prediction. Now in production, what I'm going to do is
that I'm going to pass an image the image copy
is going to be created. As we do not want to manipulate
or work on a single copy of an image because in case
of any challenge or any issues, we do not want to overwrite or lose the image
for that matter. Once this is done then
what I'm going to do is that I'm going to call
the detect phase function which in turn will help me
identify who the person is then what is going to do is that it's going to predict the
image using our face a cognizer and then it's also going
to print the confidence or how sure it is that this face matches. Now what I want you
to understand here is that if my confidence
is low 30 that is if I'm not sure that this person's
image is correct or the phrase is detected then
what happens is if it is greater than 30
is going to draw a rectangle and then it's going to write
who the person belongs to and it's going to return
this image for me on the screen as well. Now why have two images that I'm going
to be passing here? If I just go onto my desktop images show
you first is abc dot jpg, and the second is wrong dot
one dot jpeg as well. So let me just show
you the images. So you can see here. I have drawn dot
one dot jpeg, which is an image of Ronaldo. And I'm also going
to pass a b c dot jpeg which is an image of messy. Okay. So these are the two images
I want to identify and I will be passing
this to my program. Let me just run this program so that you get a full
fledged understanding of what we have written here. And you understand
how this works out as well. Okay. So once I've passed this image, then I'm going to call
the predict function so that it predicts whose face that this belongs to Two
or who's this face is as well. Once that's done. I'm going to destroy this now. I hope you've got
a simple understanding of this. If you are interested
in the code, please give us your mail ID. We will be happy to share
the code with you as well. Now. Let me just execute this so that you get a clearer
understanding of this. Now, let me just zoom in
so that you know, so you can see it has started picking up
the faces for recognition. So you can see it starting
to prepare the data. It's checking out
all the images. It's reading the faces is trying
to detect the fuses for itself. Now you can see total faces are 22 wherein
it has identified 22 images and it also has started
to predict it and it's given me the confidence
of the face detection that it has taken place as well. Now the prediction
has completed for Ronaldo. It has successfully
predicted the face and it also has put in the text
and if you see for Messi also, it has done the same
but the text is very small here because the image has been
resized for that matter. Okay, so it has successfully
predicted this it has detected. It also has recognized
it as well now. Has also told me who
this face belongs to as such for that matter. So this is how our face recognition
system works completed. Now, this is just one of a very my new door one of
the basics examples of this and if you want to create
a complete full fledged system, let me give you an idea
on that same as well. Now if you wish to have a very
great system out of something that's really effective
out there something that's not too hard
to build then this is something that you should really
consider place your camera, which is a wireless camera
on top of your door. Okay, use the Raspberry Pi
to detect the motion. What will happen here? Is that every time there is
a detection Raspberry Pi would detect this happens and then it would pass
on this information onto the cloud the reason
for doing this is that it becomes quite fast and it's way more effective in that matter as
well here the phase would be detected recognition
would happen all the information regarding these phases would
be present on the cloud itself. And once it has identified that this is a valid
user then correspondingly. It will allow the door
to be opened as well. But what you need
to understand here is At this in turn can also be configured for you to have an application or a notification sent on
to your mobile application from the cloud as well. Because most of the mobile
application Epi is for pushing information is
very easily accessible and configurable on the cloud as such the Raspberry Pi has
become the most popular platform for exploring the no House
of Internet of Things. It is redefining
the nuances for a new era of minimal operating software. And if it's
about operating systems, you can be sure to find Microsoft Windows just
around the corner. So we'll be talking
about the Windows 10 today, but not the one
on your desktops or laptops. In fact, not even the one
on your smartphones this Windows operating software
runs devices even smaller than any of those now, if you've been tinkering
with the Raspberry Pi you must try installing
the Windows 10 iot core on it and know the operating systems that come as a part
of the Windows 10 iot windows that is fairly popular amongst operating systems is
now available for Internet of Things is actually a revamp of the old operating
system Windows embedded but with iot taking the World by
storm the Windows 10 iot is now it's official operating
software for all everything's and systems. Now the windows for Internet of Things comes
in three packages the Windows 10 iot Enterprise comes
for the embedded systems that draw more power
and require more resources like the bulkier systems
in high grade or industrial grade Machinery the Windows iot
mobile Enterprise comes for more minimal systems
like the ones in our televisions or handled it. Vices and the Windows 10 iot core comes for
the smallest embedded systems in iot like even minut
combinations of microcontrollers and microprocessors. So you can scale up or scale down these options
according to iot devices and the requirement now the Windows 10 iot code
is the smallest of the three operating systems
launched by windows for The Internet
of Things it in fact is the smallest operating system that Windows has ever had
and you will be happy to know that this one comes
for free now this windows for small devices is so minimal. It's interface is unlike any of the other
Windows operating systems and it is almost like that
of a bear Windows kernel but it is just about enough to
power the things in iot run all of its applications and ensure the seamless flow
of data to and from the cloud. It also supports many other
words used in iot development, but we be focusing
on its installation on the Raspberry Pi today. Now, it's minimal interface is
for better convenience as whenever an application
gets deployed to run on the Windows iot core the
windows interface will disappear and all you'll be left
with is the application. Running on it. Now this is because the Windows
10 iot has been designed to run both headed and headless which essentially means
with and without a display. So your device becoming the application
will reduce confusion while running headless. This also means that we can have only one active
application running at a time while there could be
other background applications running all that said
I think it's finally time. We went over the installation
of the Windows 10 iot core on the Raspberry Pi so I have a Raspberry Pi
3 here with me already and a class 10 SD card of 16GB upon which Be flashing
our operating system. And please make sure that you SD card is at least
class 10 and has a capacity of at least 8 GB without these you
might face compatibility issues with the windows iot
core operating system. Now the Samsung Evo or the SanDisk Ultra
would be your best choices. So the first thing to do is
to format the SD card. I recommend using
the SD card format or software that is available
for free online. You could also use the default
format on your system, but I'd avoid using it as it
does not optimize the partition. Required by the operating system
and the storage on SD card and it compromises
on the entire capacity that is available. So go on with the SD
card for matter, so it is really easy to get
we start by opening a browser and typing SD card for matter
into it and hit enter so you can go to the first link. And scroll down
once the website opens so you can see SD memory
card format a download for Windows and Mac since we need the Windows
operating system to flash the windows iot core or to
the Raspberry Pi will be going for the windows version of it. So click on it and you
download should be initiated. So once you download
is done find the setup file and install SD card format
Iran to system. I've already done that
so I will straight away go and open the app. And here it is. So you'll have to know which Drive is your SD card
mounted on for that. We can go to my PC or this PC. So in my case,
it is the F drive. So I'll make sure that the card is wanting
to the F drive and I'll click on format. It will give you an alert saying that it is all data
on this card hit. Yes. There we go. Now we are ready for the windows
iot core download. So we go back to the browser
and we type windows iot. So the first website
is the official website for Windows 10 iot core. Once the web page loads go
to the download section and you can see there's
a download link saying download the Windows
10 iot core dashboard. So this is
what will be downloading. So we'll be installing
the operating system through this dashboard itself. Let's go to the download
folder find the setup file double-click on it. And click install. So it's a 58 Point 1 MB file. So it is getting downloaded
on the Windows 10 iot core dashboard from
where we will be installing and flashing it
onto our SD card. So once the setup is done, the windows iot dashboard
should pop up automatically if it is not you can search it
on your system as iot dashboard and open it. So now with the dashboard
in our system, we can start installing
the windows iot core onto our SD card. We can do that by setting
up a new device. Let's click on setup new device. So once you do that,
you can see a few columns that you will have to fill so
the first is device type. So I'm using a Raspberry Pi 3. So I'll choose Raspberry Pi 3. What? Was Bill do I want well, I want the Windows 10 iot core. So I'll click on that
which Drive is my SD card on its on the F drive. That's correct. And you can rename
your device from here. Let's say I'm gonna name it men win RPI and you'll have
to set a password for it. So I'm going to go
with our PI 3 win 10 have to repeat the same. Our PI 3 2010. So now that the passwords are
in place accept the license agreement and click
on download and installed. So I'll give you a pop-up
about erasing the SD card click on continue. So you downloading has now
started so you might get another pop-up which is
the iot utilities trying to install on your system, press yes, then you should get
a command line window where the OS will be flashed
upon your SD card. So with that done, we finally have the operating
system on our SD card. There will be another prompt
asking you to format disk. Please hit cancel. This is very important to not
hit for my disc it on cancel. That's it. It says you guys
d card is ready. So go back to your mypcc. Make sure you eject
your SD card. Once it is safe to remove
take it out and put it back in your Raspberry Pi. So the next thing to do is
to connect all your peripherals to your Raspberry Pi and power it on now
most operating systems will require you to configure
some basic settings. When you put them up
for the first time since the Windows 10 iot core is
the smallest version of Windows. It almost has negligible
settings to be configured. It will only ask you to set
up your default language and if you want to connect
to any Wi-Fi network, and that is it so now
upon starting the pie. You should see the windows logo. You can see the welcome screen
for the Windows iot 10 code now since it has been designed
for minimal and low-cost devices the configuration upon the first boot
has also been kept minimal so that does it for finally
booted into the settings of the windows iot core. So let us set the default
languages English and hit next after which it will ask you
to connect to a Wi-Fi network. Let us say we will skip
this step and You can also do it later
from the home screen or the desktop. So that's it. You finally successfully
booted the windows iot core and it is ready for use. So here we have Katana
you can record your voice for better speech recognition. I'll skip this for now
and let's say got it. So this is your desktop for the Windows 10 iot core you
can see there's device info which is mainly a home screen. It tells you the IP
address and everything. There's a command line window. There's also a browser
and there are tutorials for you to make the most
of the windows iot code. This is settings button. There's a power button
as well other reason behind your desktop
being so minimal is that the windows iot
has been designed to run both headed
and headless applications so to develop applications, You can do it on your laptop
and then push it remotely it onto your Pi upon which the pie
will become their application. You can use the visual
studio for it. So now you can go back
to your iot dashboard. Now the iot dashboard also
offers some functionalities for you to remotely
control your pie or your windows iot core. So you can go to the my devices
on your dashboard and you can see there's a Raspberry Pi
3 with the name mean when RPI the name that
we gave it in the start now. From here, you can open it in device portal
to our extended features. You can launch
a command line window. You can launch
iot Remote client. You can shut it down. You can restart it plus you
can add another pie to it. You can also connect it
to the Jour Cloud for intelligent analysis
and advanced cloud computing. And also there are
three packages which indirectly be deployed to your Raspberry Pi so you basically
won't need Visual Studio. You will not need much
of coding you can directly download these packages
and run them to die. The power of Windows iot core. So that's it. You can see the settings offers some basic preferences
this network in Wi-Fi. This blue to this. Katana is a command line window. There's an internet browser and there are other
tutorials to make the most of Windows 10 iot core. So that will be all
I hope you followed how to install the Windows
10 iot code now looking at the rise in the Popularity
of Internet of Things. I've come up with this tutorial
to make sure you people pick or purchase the right
development boards for your iot projects. Now the adreno and the Raspberry Pi are you go
to votes for all things iot, but how are they any different and which one
should you be using? Well, don't worry by the end
of this session. You could be educating
your friends on the same. But before we begin though, let me quickly tell you
what exactly I'll be covering in today's session. So for those of you that are still confused
on what these devices are. I'll briefly talk about
these integrated circuit board. Odes then we
compare their hardware and software specifications
understand their modes of operation check out
there available boards and finally conclude
with the purpose that each of these boats
have been designed for so without putting
it off any further. Let's get started. The computers were
a breakthrough in the world of machines making
the operation of any device or machine smarter than ever and the motherboard
in a computer is what drove these operations either performing
logical instructions that have been fed to them or calculating outcomes based
on the information. They receive these
motherboards LED all parts of the computer communicate
and act by utilizing the CPU and the memory in fact
any peripheral connecting to the computer also
needs to connect to the motherboards
programmable input/output. Actors for their operation now these motherboards
are integrated circuits and with the world trying
to minimize the size of devices and make them more portable. We now have a variety of
similarly capable circuit boards available to help us build
smart equipment of our choice and according to our needs. Only smartest development
boards today zero down to the Raspberry Pi and the Arduino
both these integrated circuit boards might have
a few things in common, but differ majorly
in the mode of operation and capabilities and if you've been experimenting
with the embedded systems, or if the onset of iot has got you playing with
these integrated circuit boards, then you must understand
the key differences between them and know the one that will be more suitable
for your iot projects. So let me Begin by telling you that the Arduino is based
on a microcontroller and is mostly designed. To control the electrical
components connected to the circuit board
in any system. Whereas the Raspberry Pi
is based on a microprocessor that computes data and produces
valuable outputs and if required it controls
the electrical components and it's system based on
the outcome of its computation. So basically the Raspberry Pi
is a single-board computer with the Arduino is just
a single-board microcontroller. So let us start
with the comparison by learning more
about the hardware and software specifications of the latest models
of both these devices. Now the Arduino boards have a very simple hardware and
software structure the Arduino, you know runs on an 8-bit AVR atmega328
microcontroller on a chip, which has a ram of two kilobytes
upon an eeprom flash memory of 32 kilobytes and clocks are processing speed
of 16 megahertz on the contrary the Raspberry Pi boards
have a complex. Lecture of hardware and software the
Raspberry Pi 3 B plus comes with a powerful
broadcom BCM to a37 b0 microprocessor on a chip as
its Central Processing Unit, which runs on a 64-bit
quad-core processor called the a arm cortex a53 featuring a static Ram
of 1GB running directly upon the SD card storage
at a clock speed of one point four gigahertz. Now this microchip also acts
as the graphics processing unit with the broadcom video
code for integrated graphics card building. It for low-power multimedia
processing now such a multi utility microchip
is being called a Soc or a system-on-a-chip these days and besides the processor
and the graphics card. They also house other
microcontrollers to control the electrical components. So there's plenty
of power packed into it. Even for many add-ons
or USB connections. Now most of the other Hardware, uh, pretty much the same
for both these devices like timers voltage
Regulators serial, peripheral interface bus has
into integrated circuit serial, you are and I/O pins, okay. Connectors but again, the I/O connectors
in Arduino are more capable than the I/O pins
in the Raspberry Pi which must depend on transistors
to drive external hardware. I'll be Raspberry Pi is still
better logical processing which means it can compute even
complex mathematical operations enabling it to process and support audios videos images
and other gu eyes. So it's more like a computer
in contrast to the Arduino which can only control
path The device and would need additional circuit software and communication buses
to run audio video or any visual graphic at all. Plus the pie also
offers HDMI support. So all that you need
is just a screen or any display device removing
the need for connecting it to a computer all the time
like the Arduino now this is because the Arduino comes without any operating software
the only real software it has is for compiling code
to Machine level and allowing serial connection
for data transfer. This is also why
it requires less power and could even get faster as it only needs
to compile code and use the chip directly
to control any component. Whereas the Raspberry Pi must enable all required
layers of software and utilize different programs to make any task
happen then again, it is also why the Pike
in multitask doing a variety of tasks at a time while the Arduino can perform
any single task multiple times. So the Arduino offers
an ID with the compiler and is the only one
that this device supports which means only the codes
written here will work and get compiled
on to its physical board this He uses basic C and C++ paradigms and breaks
down all its functions into accessible packages
making it quite easy to learn and operate however
with the Raspberry Pi which supports multiple
programming languages and software environments. You need to be proficient at coding to get
the most out of it. Also the pie now
comes with Bluetooth and Wi-Fi support
for remote connections. In fact, its latest model
features faster Ethernet or dual band of 2.4 gigahertz
and 5 gigahertz wireless, LAN and an embedded web server while the we know features
Bluetooth on just one of its boards all the Arduino BT and will require
additional circuit for communication over
wireless networks or ethernet. Now these additional circuits
for the Arduino or called Shields and are attached to it
for extended functionalities similarly in the case of the Raspberry Pi
these extended circuits are called hats both are plenty of available housing
diverse functionalities sensors Motors actuators and software on them for interacting with
the physical world and driving external hardware according to It's
host devices capabilities. Okay. Now I can see
somebody has a question. So Debbie wants to know what is the most
popular programming language that is used for coding
with the Raspberry Pi. So since the Raspberry Pi is essentially a computer
you can actually run code on it in almost
all programming languages. But the most popular one
in my opinion would be python as even the pies official
operating system comes with the default python ideally, which is the IDE for python. So I hope that clears you doubt
if there is Any more questions, please keep it coming. Okay. Meanwhile, I mean move on. So let me now tell you how you can identify the purpose
behind using either an Arduino or Raspberry Pi. So the Arduino being more
of a microcontroller motherboard is best suited for
repetitive tasks like reading and Reporting temperature
opening and closing of doors controlling lights or building motion
detecting alarms. It also works great for creating quick
and lightweight prototypes, but the Raspberry Pi is more
like a general purpose computer and is good at complex. Relations or performing
multiple tasks or processing a large amount of data. So if you need
a full-fledged computer on a very small scale the pie is
what you should be looking at. So let us now move on
to the different boards that both have had over time and look at some
of its prominent ones. So the Arduino has been around
for quite a while now and has had multiple
boards over the years with different functionalities. Although the Arduino, you know, the Arduino Mega they
Do you know lily pad and the Arduino Bluetooth
have been the more popular ones? Similarly? The Raspberry Pi despite being relatively new is
on its third generation with each of its models
better than the previous ones the Raspberry Pi 3, B+ is its latest and most
powerful development board yet in the session. We'll be talking
about the top seven projects which are outstanding
themselves in the Internet of Things domain now, let's talk about Each
of these projects. Let's start with number 7,
which is a biometric system. Now a biometric
system is something that we always encounter
on our daily basis as well because we always
either use a fingerprint sensor or we use an iris scanning. So it depends from
organization to organization. But how does this actually
work the system there? Let's take in case of this example has
a fingerprint scanner. Now the first time that I
am presenting my fingerprint, it scans the fingerprint and considers this as
part of an enrollment. Us from this fingerprint
template what it does is that it extract
certain key features, which makes it different
from others and stores it into a database from then
forward every time that I place my finger on top
of this fingerprint scanner. It creates a template and Compares this
with all the templates that's present in the database if it matches
then correspondingly, let's it gives me an attendance
or lets me access at or if it does not then
it raises an alert with respect to that same now. This is just a foundation now as I said this biometric
system Fingerprint, it can be is scanning or it can be a combination of both as well as
voice recognition system is one of the key products
in the biometric domain next. We have Smart irrigation system. Now I am someone who personally likes
gardening a lot. Now, this is something that would really
make my life easier because usually what I do is that on Sundays I spent about two hours just
watering and looking after my plants through
a smart irrigation system. What it does is that it checks the moisture
present in the environment or in the water lanes that I
have created now to help. You understand how it works. Usually there are two main
Internet of Things devices that are used here, which is the Arduino board
and the Raspberry Pi your Raspberry Pi becomes
the main processing unit. And I place an Arduino board
for each of my water channels. These addressing modes
themselves are connected to multiple sensors which are part
of this water channel. So what these sensors do is that we check the moisture
present in these lanes are such so let's say a specifically in does not meet the minimum
required moisture then what it would do is that it will Send a signal to
the Raspberry Pi again. All these devices are connected on the same
wireless router Network and the Raspberry Pi
would identify the lack of moisture and pass
the signal to the relay. The relay in turn
would initiate the water pump and water would be pumped now
in order to ensure that water is not wasted we
would create gate controls and only the gate where the moisture is less. Would the gate be open
once my sensor detects that the moisture level has gone beyond my required
level it would Against another signal to the Raspberry Pi asking
it to stop the pump as well. So this in turn helps you
to save a lot of water and also makes your life
quite easier as well. So after this your only task in your garden would be is
either setting up new plans or creating new water channels
as such the next project in our list is a security camera
and door unlock system. Now, this is something
that's quite interesting and I have personally
tried this out and it's really something that you should try
out as well here. What happens is
that you place a camera on top 40 Which in turn
clicks the photo of a person who comes into the frame now, this photo is again sent
to an analytical system which in turn Compares this
with all the photos that it possesses
in order to identify whether to let the user open
the door or not now and evolution to this is that if it does not find
the photo of that person it can notify the concerned person that so-and-so person
is trying to access this dough. Would you like to
authorize this person and address information
to the database or would you? I'd like to deny the access
to this person as well. Usually this is used in areas where you have high
sensitive information stored in order to maintain
a strict control to the access to this information
as well another usage of the security camera and door unlock system
can be even at our homes when we can identify who's come at our home
when you're not there and either decide to give them
access to our homes or not. Now the next thing is something that we all really
decide to have which is a smart home system
a smart home system can be something
that really makes. Life quite easy starting
from energy management where the light control system
is the AC the appliances that we use the thermostat
all this is managed in short trying to cut
down the power consumption that's taking place by door management system
is also part of this. My security system is also part of this my water management
system again becomes a part of this as well. Again, these are key things that really stands out
in the smart home system. But again, what I
would personally recommend is that a Smart Homes
limitations is where our imagination stops anything that you wish to automate
or make wish to make your life easier can be part
of a smart phone system as well. Now a smart home usually
is going to be a base for our next project which is a smart City
a smart city is an evolution of a smart home here. It's not just the sensors
of a single home that is connected here. It's correlation or a network or a connection
between various organizations various domains as well
as various segments of that city as a whole
as such Judo the life of every single dependent person in that City becomes easier
as a whole as well and in turn will really
help develop that City to Greater extent such now
the key factor here for a smart city is
a government support as well. And if governments are
really willing to take this step then I hope we would see
a smart City completely built on internet of things maybe
the next 5 to 10 years as well. Now the next
project is something that really stands out
on a personal level this this is Zelda Ocarina controlled
home automation system. This is personally something that I feel is the closest
to a smart home system where most of the elements of the home is completely
controlled by an Ocarina what Alan pan has done here is that he's created
a node-based recognition system, which completely automates
his hope but rather than telling more about this. Let's just look at a quick video which will give you a glimpse
into how he has done this and what has he done. 7:27 p.m. Next on our list at number one. If you've not already guessed is Jarvis Jarvis is the
artificial intelligence system that Facebook's creator
Mark second Berg has built for his home automated system. I'm quite sure you've
already seen the video of Mars are kannberg interacting with Jarvis which has
Morgan Freeman's voice. And if you've not, here's a quick glimpse
into the same. Good morning, Jarvis. Good morning Mark. It's Saturday. So you only have five meetings
room temperature is set to a cool 68 degrees
earlier this year. I started building a simple AI
to help run our home. I talked to Jarvis
using this app. I built it uses artificial
intelligence to understand me and figure out what to do
Max woke up a few minutes ago. I'm entertaining her. All right. Let's go check on her. Good morning. Let's practice our man. Zhao Xiang Hao Jarvis
your Mandarin is so soothing she she Jarvis also helps
me get ready in the morning fresh shirt Byron home. Hell yeah, Jarvis knows when to
make me breakfast. Your toast is ready. All right, it's time
for my call with shrap. Can you get him
on the video conference line setting up the BC room. Now remember to check on the AI
guidance system for tequila. One of the best things about
Jarvis is it could recognize people at the gate let
them in automatically and then just tell me about it Mark your parents are coming in. Thanks Geoffrey. It's Jarvis and Jarvis can play
all of our favorite music. Hey play some good. Go back songs. I'm sorry Mark. I'm afraid I can't do that. There are no good
Nickelback song Good that was actually a test. Okay. How about just play some songs that are whole
family likes what? I'm a dad. Now. This is what I listen to. No, what you need
to understand here is that internet of things
is not something that's just dependent
on a sensor or a few sensors, which is connected
to a Raspberry Pi on adreno put when I look at the entire
architecture of iot. This is a complete ecosystem where in my sensors
gather information, which is again stored
on a platform and then processed on this now if there have been
any issues or failures that have been reported
by these sensors, then I need to create
actionable items. In order to ensure that this is never
really repeated again, but the information
wants to spin process the process data is then passed
into machine learning and artificial intelligence
in order to understand analyze and identify various patterns that's out there and help
pass this information on back to the sensor
this in turn helps. You have a better experience and also improve the system
as a whole a system where I come home every day at 8:30 start off
my Casey wait 15 minutes for it to cool down is
a tedious process. But it's a today my Internet
of Things platform identifies this pattern already switches
on the AC at 8:15. Then I just need to come home
and I can relax right away. This is just one
of the key examples of one of the key ideas that's out there today and the limitation here again
is just your imagination. It is so good to see that so many of you
are keeping up with the trends. Of technology today. So let us dig deeper
into these devices that Mark the beginning
of every iot ecosystem and look at the best ones in use
today in different domains. But before we start
don't let me quickly run you along the outlines
of today's session so that you have
a clear idea on the topics that I'd be covering. So first I'll be telling you
how you can identify any object to be an iot device. Then I'll tell you what they
can do for you and cite some of the major use cases on it
and then I'll tell you how iot devices are shipping
the entire world and code some leading examples and then I'll tell you
how secure is. The device in iot and finally, I'll be telling you all
the important things that you need to know
to build an iot device. So without putting
it off any further, let us begin our session today. Now these devices that play the role
of things in iot are what make the smartest
systems possible today be just a smart Appliance
or an entire Smart City. It could be complete
Computing devices designed keeping the portability in mind or tiny electronic
components operating on minimal device software
for data transfer and Activity without
any Computing abilities at all and there are also objects that are not made up of
any electronic Hardware at all and can still become
the things in iot by having these smart things attached
to them in summer the other way. So what I'm basically saying is that even we humans
can become a thing in the internet of things. But again, that is a contextual top
and you'll understand why as we move on with
the session today. So for now we can just say that any device however puny or powerless with
the ability to receive or send data allowing
it to communicate over networks can be called
a thing or an iot. Ice also, it should be serving
a purpose like either collecting some information or
delivering some outcome. Now that is the least
any iot enabled device or thing in iot
will require today, even though the bigger devices
have diverse features and can be more powerful. It is a miniaturizing
of computer hardware that has truly given
the Internet of Things its Edge in the market today. It is made it possible for these iot devices to even
function and microscopic levels in the remotest corners
of our world. And as also Borden focus on more Target oriented features
specific to their goals, Plus with devices running
on incredibly low power levels and operating with
such minimal resources. The things in iot come
with the promise of implementing Solutions in the most cost effective way
possible without compromising on the system's accuracy. This is because at the smallest
levels these things are embedded with just enough technology for them to communicate
over the internet and other networks to get
the data analyzed stored and processed over that Network or over the cloud rather
than physically housing. All those Technologies
on themselves. Also such low power levels
are perfect for the lead. Our chips sensors
and other low energy components that is devices used
for the operation. But like I said, there are substantial equipped
to connect wirelessly for exchanging data and Performing
actions based on them. Also such low energy modes
of connectivity make it easy for these things to operate
just about anywhere while letting us control
and monitor them remotely for almost any part
of the world. So it is as if the British technology Pioneer Kevin Ashton
who coined the phrase internet of things back in the 1999
somehow knew of these potentials that I would have brings out. Devices today so
let us move on now. Let us see what things
can do for you. So one of the most popular
implementation of these things and their applications
have been in the sector of home automation today, the use of Internet of Things in this industry
has got the world closer to achieving a dream home
powered by the smartest systems using these iot devices and these systems
have been built such that all its devices can intercommunicate to allow
owners a customized access. To all aspects of your home like your lights your logs
or the inside environment or security cameras or even your total energy
consumption with the brand specific IOD platforms
and clouds behind them. These iot devices are
now capable of exhibiting ambient intelligence and also stay connected
to the internet for driving real-time operations. So let us get into some examples of iot devices used
in most Smart Homes today, and we must begin with the
tunable smart Lighting systems like Philips Hue
smart home lighting or the tp-link. Color Smart Wi-Fi that can be easily controlled using your smart phones
with the option of creating custom
lighting automations from anywhere in the world
by just being connected to the internet
with smart lights like these you could turn
them on or off anytime and tune the color and the brightness of Lights
in your home's remotely no matter wherever
you are moving on. Did you know that locks and keys are getting
old fashion nowadays. It's not a joke. Smart locks are a part
of almost every smart home today and the August smartlock
third generation is one of the newest and the phones
on the market today. What makes it Stand Out
Among its peers is that it comes
with bluetooth support so that you won't need to worry, even if your internet is down
somehow with a device like this that attaches to your existing
that bolts you will always know if your door is
completely closed and locked with its doors and Technology. It'll even remind you
to lock your doors when you leaving or tell you if they were properly closed
or not based on the location of your phone, you can even create
secure virtual or temporary keys for your neighbors and guess just in case now the next iot device is Nest
learning thermostat it is one of the best examples
of an iot device today as a part of your home automation system this intelligent device
Lawns your schedules and preferences over
the first couple of weeks too smart in the cooling
and heating systems in your home and then regulate
your home temperature automatically to save
electricity by 20% among a ton of other features. So to reduce your energy bills, it uses sensors and your phone's location
to realize the whether the time or the environment inside and when nobody's home it
will shift into energy saving. Mode apart from the
thermostat Nest also provide smart cameras with smoke and carbon monoxide alarms to better manage
your home's atmosphere and Security Plus it
lets you manage and monitor each of them
through its dedicated app or even build your own
Revenue channels upon them. There is also one more device
that I think is worth the mention in the home
automation systems of today. It's mostly just a doorbell
for your front doors. Only a lot smarter
the video dogus by ring is a modest name
for all the things that it can do. It's like Always home as it lets you answer
your door from just about anywhere it will in fact, even a lot you of motion
within 30 feet of your door and stream its live footage
onto your phone. And if somebody rings
a bell you can let them know that you're not home and even have a word with them with its right Vision the wide-angle high-definition
recordings get simultaneously saved were Cloud
for your use anytime. Now before I tell you what else iot devices can do for
you virtual or home assistance like the Amazon Alexa
the Google Assistant Siri and Jibo are also attending
application of the internet of things and have made it big in the home automation
systems of today. Now, other than making
your Home Smart iot devices are also capable of intelligent Lee
tracking your health and fitness and as far as Fitness
goes the Fitbit tracker or the Apple watch and other such smart watches
have been a revolutionary success and along
with your Fitness aspects. They also solve your day
to day Health Care needs like checking your heart rate
and perspiration levels and tracking your body
temperature to predict a cold or a flu on the Are you
will be amazed to know how the other iot devices
are saving more lives and making Health Care better like the Philips Healthcare is
medication dispensing service or the hero smart pills
dispenser are the most successful use cases today and exemplify the potential
of iot in the healthcare sector. It is the perfect reminder
for your medications and is a convenient measure
for the elderly and the sick alike with the option to remotely
manage it with your smartphone. You could also
monitor the medication of your near and dear ones
and be notified upon running low on any Medication at all plus you could also get info
on the right medication for common health hazards. And now with how integral smart phones
and internet have become in our lives Health and Fitness Solutions
using iot will only get more and more prominent
in the recent years to come. So apart from
the personal benefits that you can get as a consumer
or user iot also aims to impact the greater good the talk of smart cities
is prevalent these days where the waste management
system is more efficient. You shall in the environment
is checked and even the outdoor lighting and traffic signals are equipped with motion sensors
to ensure energy conservation by turning on only
upon detecting traffic updates and its route or upon sensing pedestrians
and Vehicles nearby. Now, let us look at some
of the other sectors of service and Civic amenities that these iot devices are
powering to up the efficiencies of our world today. One of the most visible
Smart Systems in place is in the domain of transport
and Automotive today and along with more and more
connected car platforms today there has Also
been a lot of hype around smart and self-driving
Vehicles hitting the lines of Transport very
soon plus with the maps providing real-time Intel and the modern vehicles housing
numerous sensors in them. You could always tell
which way to go or what parts need to be looked at. Like say tracking the Fulton's
option altitude and maintenance issues of flights
in real time without waiting for it to land every time
can help in anticipating problems to schedule maintenance
prior to its arrival. So that deal is
and mishaps are minimized in the aviation industry
another major area of Application for
iot devices isn't the field of Agriculture or sector that is often neglected. Despite. Its utmost importance can now
be brought up to speed with iot with several cheap
and minimal sensors that monitor the best climate
and soil quality for the right kind of crops
or other smart devices that ensure the efficiency
of automated irrigation systems, even the gardens in your home
or the trees in smart cities or even the plant life
and vegetation in agroforestry or wildlife habitats could be kept in check the most
Innovative iot devices to they include the smart
watering system Blossom, which can create
optimal wanting she Duels for all the plants in your home
based on real-time weather data and forecasts and will regulate
all your sprinkles accordingly and allows you complete control
over them through Bluetooth or the internet and also the clean grows carbon nanotube probe is one of
the best iot devices for farmers and gardeners all around the world today with
sensors to monitor the intake of nutrients in the crops to
better manage farming resources and improve the quality
of their fields farmers can now order Maturity rate and the color of props
for better yields and faster. It's a production now. It might be a little ironical
of me to talk about one of the very first Industries
to be made smart on such a late note
in today's session, but I really just wanted
to save the best for last and believe me the sector
of retail and Logistics is where the internal things
promises the most astounding results with iot devices already
being extensively used in shopping restaurants
hospitality industries and many other
businesses to control the supply chain effectively. And obtain valuable
insights based on them and manage their logical
or merchandising expenses in the best possible ways. Now the Q hop is one of the leading examples
of internal things being used in the retail today typically designed to bring in
seamless autonomous check out Technologies to
all retail verticals by digitizing the checkouts
through RFID tags that only unlock
after its payment is processed. The sole purpose behind it was to allow users
to sell check out in stores. Although it is mainly used
for security reasons today to inhibit Petty thefts. Doors, but guess what that was before the Inception
of Amazon go earlier this year. Now the technology inside this convenience store
will really seem like it was pulled out
from the future somewhere. Although it is so new that there are
only three locations with this feature
has six toes so far. So operate and are managed by
the online retailer Amazon. These stores can give
you a shopping experience unlike any you've ever had before with the idea
essentially being grab-and-go. These stores will just
need your Amazon go apps for you to enter
and then employees. Eurovision machine learning and Sensor Fusion
to automatically add items that you pick instantly on to your virtual card
will also remove them of just as promptly if you
keep them back now once you've grabbed
all the things on your shopping list and are
ready to exit all you need to do is just walk right
out of the store. Yes, no more lining up
in busy cues of the usually limited checkout counters to
wait your turn for the purchase. You can exit the store without even having
to pull out any cash or a card and you will find
the amount for all the things that Walk out with the beaded
simultaneously from the balance in your Amazon account such
a smart system could also check the inventory regularly
to notify retailers on the need for restocking and even helps to manage
the supply chain and a better way. Now, there are also other
promising areas in logistic that iot devices have started
to unpack like in the case of shipping cargo or Fleet
Management smart Bluetooth. Low energy tags are attached
to the items being moved for remotely tracking the exact
locations speed of transport and storage conditions. For instance. The things dot IO is
a simple iot platform that provides a dedicated
cloud-based dashboard for better Logistics
by enabling access to real-time and reliable inputs from it's
connected smart sensors and paired location trackers
irrespective of wherever they might be so although use cases and the respective
iot devices are things. We just went over
our only a select few of the countless applications
of the internet of things that can drive almost
any sector today and these iot devices
could be things themselves or even attached to someone or something to make
it a thing in iot. Like even a person with a heart rate monitor
could be a thing in iot as it collects
and provides information that can act as inputs to other Smart Systems
to operate on so the Baseline for anything to become
an iot device could be laid out as any object with a unique IP address
for communication over networks and the ability to gather
and transmit data or receive data and perform tasks based on it and the Ameri technology
in these devices are what interact with
either the internal States or the external environment. To capture all data and drive decisions
made up on them. Now with the internet
of things being such a large community of different
devices the challenges of iot devices mainly start with
communication as the protocols and languages used
by each of them vary. Hugely due to the lack
of common standards for all of them yet and this lack of a uniform and secure standard
across all iot devices, which is great security risks making them highly unreliable
for most important. Missions are into
operations today. Also without customers
being assured of the privacy and security of the data. There is just no reason
for them to risk using or adopting such
insecure Solutions. Do you really understand
what the role of security and iot devices is today. Well, let me tell you
about a major Cyber attack that happened back in October of
2016 a large distributed denial of service attack dubbed
Mirai affected DNS servers on the east coast
of the United States which disrupted Services
all across the world upon further investigations. This issue was tracked
back to the hackers that infiltrated smart networks through the iot devices
being used in them like the routers or the camera. So that brings
about the situation. We're in our devices and data
are all remotely connected and stowed upon these networks and its security
gets compromised. We might even not know it
over the first few days or weeks and it might be just
too late by the time we do realize now the way in for these hackers were
undoubtedly a result of poor practice at some end like say the use of default
passwords rather than changing. Hence, the adoption
of better practices and the reinforcement of proper authentication
Network segmentations encryption and cryptography can
still make the things and its systems
quite secure given that we start making sure
of building them up securely from our end as well plus the issue of overall
connectivity is also a feat that our world is still
driving to achieve but hasn't been able
to accomplish it. That brings us
to the last topic for today. So let me tell you
the important things that you should know
to build an iot device. So building an iot product or device or solution
must be done thinking about the relevant purposes that it can serve and the ways in which they
can be prepared to work for at least the next couple
of decades along with the option and space for
quick improvisations and upgrades and just like we saw in all
our examples today. You must have understood
the two important categories that iot devices
are mainly made up of the first is the hardware
as And the underlying room is to aggregate the hardware
in the most minimal way. You can without compromising
on the primary features that you want
your device to use. Now. These features are mostly due
to Bluetooth low energy sensors or beacons connected
to the internet or a customized product with
probably a combination of these on a circuit board made up
of a semiconductor like Silicon and may also house
or the components like transistors resistors
receivers transmitters actuators and integrated circuit
or a microchip. So the most devices
like a smartphone or a result of such combinations
on a little more complex level with proper casing and with the second
most important aspect of these devices. So the second important category in iot devices is
the software aspect on the device level the size
of your software will depend on how minimal or bulky your device is. And what are the components that your housing on it
on the most basic note. The device software only needs to be enough
for handling the operation of your device driving
components to collect data and converting them
into transmitters. Fun connecting two networks
driving the transmitter to send data and the receiver to receive
data to and from the network and converting it to the machine
understandable form for driving components to perform
some task or display. Some outcomes based
on the received instructions. Now all this will require
very minimal software unlike the software
on your phones that come with entire
mobile operating systems all the other important
software for intelligence and smartness in these devices would be provided by the
underlying Cloud infrastructure and even by the Mobile apps
or web dashboards. The software aspect
of iot devices is in fact what controls the hardware
aspects to sense some information or
perform an instruction while both aspects are the most
integral parts of an iot device. You might also need
to overcome a few more hurdles before you iot device
is truly ready to implement smart Solutions and Power Smart Systems
of our world today. These hurdles could
be connectivity issues or compatibility problems or
security and privacy concerns, but don't let
the obstacles dishearten you in Just consult people
with knowledge on the same. While the Internet of Things
equips a multitude of domains and millions of devices
with connectivity every day. There are still a few domains, which I find a tad
more interesting than the others and they are iot in everyday
lives iot in healthcare in smart cities agriculture
Industrial Automation and finally in
disaster management first, we have Internet
of things in our daily lives. This is probably a Hallmark
in the iot industry as it's also One of the first industry is
to deploy iot at its service. So let me give you an example of how iot can service
in our daily lives consider a home appliance
such as your AC currently, what you do is that you go home you
turn on your AC and wait for it to reach a temperature that you like say about
25 degrees Celsius. So does anybody
see a problem here? No one that's probably because there isn't
a problem here. This is a perfectly
functional setting. But what if it could be better what if when your car
was five minutes away your AC received a message. What if it was
connected to a cloud which had a dashboard containing
all the relevant information like the location of your car
the outside temperature and the temperature at which you liked your room
your AC could then turn on before you arrived
and create an Ambience that you like? Wouldn't that be something? Well, you can remove
all the waters from the previous scenario
as this no longer. Must just be a figment
of your imagination. Thanks to the internet of things
iot can connect your fitbit's to your vehicles
from your smart phones to your in-flight services from
home appliances like your ACS to whole entire cities. Maybe this is what Kevin Ashton meant
when he spoke of this concept where every area sensor man
and machine could be connected to one another. So I hope you all are
with me till here. So next we have
iot in healthcare. I'm sure most of you have heard about the Smart Medicine
dispensers by now as the name suggests. It's basically a smart appliance that stores dispenses and
manages your medicines for you. Now. This is a very small piece of a very big picture
the health care and the general practice of medicine majorly
faces issues in one or more of these three things. We have your research
devices and care. Medical research has to rely
upon leftover data in control settings
for medical examination. It lacks real world data, which can solve
critical conditions iot could be the answer to all
these problems The Internet of Things opens ways to a sea of valuable
data through analysis and real-time testing Internet
of Things empowers Healthcare professionals and
improves the quality of care. Finally It ultimately
reduces the unsustainable. A high costs of medical devices. So here is an example. This is basically an outline
of how a care device Works a care device
has certain parameters that are considered safe. Once one of them is breached the sensor immediately
realize this message via a secure gateway
to a cloud now. It is vital for this
gateway to be secure as it holds all your valuable
medical records the cloud then passes a remote signal
to a Mod device that is monitored
either by a nurse or a caretaker at home. The beauty of remote
patient monitoring is that patients can now
replace a long wait at the doctor's office
with a quick check-in data share and instructions on
how to proceed iot. Hence Bridges the gap
between reading devices and delivering Healthcare by creating systems rather than
just equipment with that said, let's move on to one
of the most talked about prospects of iot
the smart cities. Now the thing
about a smart city. Is that a smart City solution is
very specific to that one city. The problems faced
in Bombay are very different than the problems
faced in Delhi. Even Global issues such as Waste Control
traffic management availability of finite drinking water housing and pollution impact different cities with
different intensities. So the only way to really
make a city smarter. Is to cater specifically
to its problem one such problem consistent among most
Urban cities is traffic. So imagine an intelligent device like a traffic camera a camera
that can monitor the road of traffic jams
accidents rains Etc and communicate that status
to a Gateway. Now this Gateway also receives data from various
other cameras from all around the city this in itself could form
a Citywide traffic system. Now where can we use it? So let's say the municipal
Corporation has decided to repair a road which connects to a highway
for more than one city. Now this could cause a massive
congestion a smart camera could send this insight to the Citywide traffic system
now considering this is a smart traffic system
it quickly learns and predicts
patterns in traffic. It will analyze the situation
predict its impact and relay this information. Two other cities that connect to the highway via
their own monitoring systems. Now the traffic
Management Systems, then can derive routes
for cars around this projects live instructions could be sent
to drivers via radio channels and their respective
smart devices. What is a bonus is that if there is a school
or an office building which has no other way but to use that road
under construction, they can be automatically Ali
called and asked to reschedule this creates a network
of self dependent systems which leverage real-time control and this is just one example
of the potential benefits of iot applications. So next we have a sector which is most neglected
despite the importance. It holds that is agriculture. However manual handling
often results in loss of energy labor cost
and other inaccuracies, which may call its Is
less effective internet of things here can provide with the number
of solutions Precision farming smart irrigation and smart Greenhouse
are a few to name. The first two cases are
pretty similar as in both cases. There are sensors
detecting various parameters at each level of the soil. We have moisture
content temperature and weather conditions one can tell the correct
depth to so the crops or the right time to water them, but One of the more
intriguing Solutions is the Smart Greenhouse a lot
of you might be wondering what is a greenhouse. So the greenhouse is
a farming technique where you can increase yield by
controlling natural parameters. So if we could use embedded
devices in these greenhouses, we could not only just
monitor it regularly, but we could practically
control the climate inside the greenhouse. So how the system works is that the sensor sends
various Toes and areas of issue inside the greenhouse. They then relay the information
via a connection gateway to the cloud which then sends remote commands via the same Gateway
back to the sensors. So once these signals
are acknowledged they are passed on to the switch gear which then activate
the lighting pumps and turbines inside the greenhouse which then create artificial
sunlight and sometimes artificial precipitation. Inside the greenhouse what more
from the connection Gateway as you can see, there is a monitor through which 24/7 real time
monitoring has come into effect. Also, you could use any smart
device to remove the conditions inside your Greenhouse. So with elimination
of irregularities and human errors, this creates a much
more efficient system. So with that we conclude
the role of Internet of Things in agriculture. So next we have another domain where iot could prove
to be a game-changer? This is one of those fields where both faster developments
as well as the quality of products are the critical factors for
a higher return on investment. I'm talking about the
Industrial Automation sector, but there are still
a few problems that need to be addressed in this sector. I think you all might
have already guessed. What I'm going to say. Well I OT comes to the rescue
Again The Internet of Things improves the line
of command immensely. It optimizes packaging and makes quality tests
so much easier to run what more you do not even have
to worry about the training cost of too much staff or a lot
of stuff going on a holiday because these factories
pretty much run on their own you can monitor
the supply chain in Real time while keeping an eye
on your inventories with iot applications one could
even re-engineer products and their packaging to deliver
better performance in both cost and customer experience. So this brings me to the final
application for today that is disaster management. Now the iot cannot stop
disasters from happening, but it surely can help in preparedness and resilience
during one due to high. High population density
poor evacuation infrastructure and exposure to severe weather
events developing countries are more exposed to the risks
of natural disasters and often have limited means
to sustain the effects as a consequence according to a study
more than 95 percent of all deaths caused
by a disaster occur in developing countries
iot can compensate for this by prediction
preparedness response. And Recovery to rescue
developing and emerging countries from their
vulnerable positions. Let me give you an example
of a forest fire where iot could be used as
a solution to be prepared for critical incidents. Like a forest fire sensors are
installed near the perimeters of the forest. They contribute to data feeds about temperature and carbon
emissions to a control room, which is common
to the entire town. This is done
via a secure network. Gateway in real time, they anticipate problems
and put defenses in place to mitigate
that impact namely the police the nearby hospitals and most importantly
the fire brigade if such an emergency occurs
near urban areas people and communities are alerted via social media conventional
media channels and SMS. This brings me to the end
of all applications that I had listed now
that we've gone through a few. Applications of iot let's move
on to its future scope in just one year internet-connected devices went
from five million to billions, but there are still a few areas
that need work. So the device is forming
the base of iot Are Wireless in nature and recite
at very remote places where energy is
a very vital issue. So with growing functionality
of each device, we need algorithms and Hardware
that are energy efficient. To avoid quick
draining of batteries. We need to make sure that sensors are active
for longer durations, like any other
advancement in technology, even in iot security
is a standout issue. This issue keeps getting bigger with more and more
devices being connected to one another we need
information seclusion methods to benefit end clients and secure that
data and privacy. No, it will be really tough
to implement the anytime concept of iot in reality, but this needs continuous work and the closest we
can get to this is only by reducing the complexity
of each existing time systems. We need continuous work
to reduce the gap between near real-time and actual real-time iot
applications promise to bring immense value into our lives. In fact the number of Openings
for iot professionals is at an all-time high. So I think it is
the best time to begin exploring the true potential of
this technology from smartphones to Smart cards to that Fitbit
on your wrist iot represents millions of rupees
of potential for an industry as The Internet of Things
becomes more important for companies of all sizes. It professionals are beginning
to seek out roles. Related to this growing Niche. So before going
into the session, let's quickly look
at our agenda. So we'll start off
by discussing a little bit about the current scenario in iot market trends
current salary situation and projections in iot and how it compares to other sectors after which
we shall see the salary Trends in iot by both experienced
and functional areas. Also I shall be talking
about the companies that are hiring for iot
as well as the roles they offer then I shall Discuss
the skill set you must acquire before applying into the sector. And finally we will see
an outline of the training provided a Teddy Rekha. So without Much Ado, let's get straight
into the module. So with the proliferation
of connected devices iot has ventured successfully across
all working areas in India and abroad India
is quickly becoming one of the largest hubs for iot across the globe companies
are viewing iot as a key enabler in various fields and are Come to integrating it
into existing infrastructure as the adoption
of iot has increased iot Market is also expected
to increase quite exponentially. So with that let's talk a little bit about the salary
Trends and projections. Now the median salary of iot professionals is
15 Point 2 lakhs per annum which is according
to the 2017 study this figure compared to the 8.65 lakh
per annum in the eye. It sector gives about 76 percent more to iot professionals iot
being a relatively new entrant Professionals in this area
are commanding higher salaries. This number is predicted
to go up as much as Seventeen point five lakh by 2019. So let's see how iot does compared
to other contemporaries? Like I said before I
OT pay is 76 percent more than the it sector
interestingly only 33% of iot. Specials have salaries
under 6 lakh per annum as compared to 58% in the it sector
also we can see around 16 to 18% of iot Professionals in India
earn in a salary bracket of 10 to 25 lakhs
per annum almost 5% of iot professionals occupy. The largest salary bracket
that is 50 plus LP a while. In other domains barely two percent of the workforce
commands this figure with That said let's look
at the salary trends for it Professionals in India and the US according
to the current Financial year. So what I've done
is I've divided this into two parts salary is
according to functional areas and then according to
the experience level first up. We have salary Trends
by functional area. Now as you can see, I've divided this
into four functional areas, which majorly command
the iot industry one is embedded systems
next system programming then we have the Engineering
design and finally application programming
in one end of the spectrum. We see application programming
being the highest paying job in India with an average
of 13 lakhs per annum whereas its us counterpart pays about sixty nine
thousand dollars per annum on the other side
of the spectrum. We have embedded Technologies, which pays an average
of eight point two lakh per annum in India,
but in u.s. It is the highest paid
in the sector with an average of $9,000 per annum
the system programming and engineering design sectors lie somewhere in between
with 11 point 6 lakh per annum and 10 lakh per annum respectively as for the
experience level more than 90% of freshers in the iot field
fall under the 6 lakh per annum salary bracket, which is not a bad
number for a fresher on the other end somebody who has a decade of work X
in this field can quote up to a 30 + L. Your salary now. What is interesting here is that from each experience level
bracket to a higher one can lead to almost 50 percent jump
in the salary with this. I think we've covered
the salary aspect of this industry. Let's move on
to the career aspect of iot. I'm sure a lot
of you must be waiting for me to disclose
the names of companies and the roles they offer. So the concept of iot
has given rise to a new era of economic growth and companies are Looking
at it to transform their businesses now here
you can see two types of companies firstly there are
companies like Dell and IBM which have a vested interest
their core products and services are built
around delivering and facilitating iot, but there are also companies
like Verizon and Accenture where it is more
about preparing for the future. Although iot hasn't completely
changed the face of it. It has definitely
created new opportunities for job seekers in the markets. I Be elaborating on this
in the following sections of the module which brings me to current roles
for iot professional. Now if you were to ask me
what field do you need to be in to land a job in iot the answer
would be many iot works on many layers software
plays a key role in usability and functionality Network layers are key to infrastructure and
Hardware defines capabilities and development opportunities
involved in an iot solution. Existing professionals
with transferable skills will definitely find progression
opportunities with iot. The iot is based
off of many technologies that it professionals
are already familiar with IP experts both software and Hardware engineers. And even UI designers could find
themselves to fit in somewhere in the industry. These companies could be ranging
from startups to Tech giants that are driving the industry, but with various domains that have adopted
the potential usage of industrial IO It has led
to a massive job creation. But unlike the other fields
iot job market is a complex one given the wide variety
of skill sets required in this place. Here are some emerging job
profiles for iot professionals a few years down
the line first up. We have solution architects. Now, what a solution
architect does is that it takes an idea and converts it into a design
and then takes the design and converts it
into code now a solution. Architect may not be
a back-end developer, but it takes the requirements
of the company and then turns it into solution blueprints which brings me to another very important player
in this end-to-end execution, which is the product manager the responsibilities of
a product manager may be many but mostly it revolves
around making sure that the clients get
the solution they asked for on time moving on now when we talk about iot we talk
about huge amounts of data. And where there is
huge amounts of data, we have analytics on top of it. So this is a job profile
created specifically for the iot Industrial Field, which is the
industrial data scientist. These are data scientists that basically work
with sensor data and have expertise on it. Which brings me to another very important
industrial job profile, which is the
industrial Engineers. Now as I've mentioned before Hardware plays a key role
in the development. Moment of connected
devices for that. We need material Specialists
who can work on semiconductors that are used
to make microcontrollers and microprocessors
in those connected devices. Next up. We have another role that is created specifically
for the iot field. Now iot has pushed
the developers to go beyond a full stack. We need developers
at the back end who are not only proficient in software coding but So in
Hardware programming, hence, we have the full stack developer which brings me to our final
emerging role in iot as we have all known
and predicted very soon. Our shop floors are going
to be taken over by robots. So what do we need? We need a supervisor who can take care just in case
there is a miscalculation or a breakdown. This has created
the final job profile that I'm talking about today
the robot coordinator. With that said let's move
on to our next topic. So while iot is poised
to be the next great boom in the field of jobs lack of skill is citing the biggest
barrier for companies that are looking to implement
this technology to be considered for any of the rules mentioned
earlier the qualifications. You need are no different than an IT development
roll Proficiency in coding and fundamental in object design
are at least required for your resume to be looked But the skills you need here are
way more than that businesses who engage in iot
Technologies are businesses that are invested in the future. They are seeking
forward-thinking professionals who not only meet
standards in academics, but also have soft skills and Innovative thinking I
am talking about people who can integrate deep knowledge
in embedded Technologies and concepts of cloud computing
and Edge Computing. These people are tipped
most Be in demand next up. We have networking
and communication protocol. These are very important as devices need to be connected
to one another at all times in real time. Finally. We need professionals who can convey a complex idea in simple ways either speech
written communication or other abstract methods. So the iot job market
is the perfect platform for job Seekers
to Showcase their skills and for companies
to establish relationships with These professionals so this is the best time to get
into this technology and who better to guide you then
the iot trainers at Eddie Rekha. So Ed Eureka provides
an iot certification training and the outline of a syllabus
look something like this. This training course
is tailored specifically by industry experts and a takes you through Concepts
such as iot framework ecosystem and solution architecture it also teaches you networking
protocols and application. Our in-depth you
shall be working with Raspberry Pi and sends hats and will gain hands-on
experience with Azure iot Hub and Alexa voice. This course also
has a real life demo and case studies which will help you master
the iot technology. So concluding I
would like to say that iot is amidst an explosion
this definitely suggest optimistic future for
professionals looking to dive into this industry. So grab this opportunity
while you still have it. Thank you and have
a great day ahead. I hope you have enjoyed
listening to this video. Please be kind enough to like it and you can comment any
of your doubts and queries and we will reply them at the earliest do look out
for more videos in our playlist And subscribe to any Rica
channel to learn more. Happy learning.
